ChatGPT

Day two of TechEx North America has been more of a deeper, critical examination of AI in the enterprise, but with a optimistic bent. The AI and Big Data programme opened with reference to what was termed the “AI graveyard” – that is, AI projects that seem to perform well in pilot, but don’t seem to cut it in the real world. Despite the presence of what might be a negative term, multiple speakers and sessions addressed ways in which the forward-thinking business might not ever have to experience the technological cemetery. The different show tracks of the second day of this event dived deeper into the pervasive issues that may be affecting AI deployments. Sessions in the Enterprise AI Implementation, ROI and Adoption tracks took stalled pilots as a starting point, and tried to ascertain the reasons behind faltering projects. There was a good deal of sound advice for organisations, with sessions on focusing agentic AI on specific business areas, building agent-ready data foundations (planning for success under the hood), and the realities of token-based AI charging on the business’s finances. At an infra level, there were deeper discussions too on whether companies should buy or build physical infrastructure for their AI projects, and the best ways to create durable ROI on data and AI projects when all the many effecting factors are given due consideration.. In projects where AI roll-outs get stuck, the core issue could be epitomised by the concept of the ‘personal copilot’. This works well on a single worker’s desk and for their individual workflows, but doesn’t really scale to a whole department – never mind a whole business. Many companies report having the budget to start such AI experiments at the level of the single user, and there are usually great results. When said user is a C-suite executive, a personally-achieved efficiency tends to increase the levels of excitement around the company, which has to be considered a positive. But transitioning from this point to meaningful change across the business is where many organisations find their individual struggles and roadblocks. Here was the meat and gravy of day two’s activities on the show floor and the numerous stages at the San Jose McEnery Convention Center. Cyber issues Despite the use of terms like ‘stalled’ and ‘difficult to scale’, in the Cyber Security and Cloud Expo stage, speakers cited the the speed at which businesses and organisations adopt agentic AI systems as a cause of a ‘velocity gap’. Where AI deployments are successful, they gain traction fast! But security and governance issues crop up when business units adopt generative AI faster than the security team can govern and ensure the enterprise’s safety. Like the proverbial double-edged sword, AI can be considered as a force that changes and can improve both attack and defence in the cybersecurity space. There are the issues created internally by unbounded agents and large language models, plus the addition to attackers’ arsenals of AI scanning tools that can identify potential exploits. Also prevalent among the round-table discussions and keynote speeches was the older theme of shadow IT, now presenting in its new guise as shadow AI. If staff place sensitive material into unsanctioned tools for example, or if approved AI systems are poorly bounded and managed, then the attack surface can expand without the cybersecurity team even being aware of it happening. Therefore, data governance and system oversight are becoming more intertwined than before – this was the message from both cybersecurity strands of the show, and the Cloud and Big Data elements too. For pure-play cybersecurity functions, zero trust was presented as one answer to the runaway adoption of AI outside the auspices of cybersecurity teams – the adoption the ‘denial by default’ position for humans and machines alike. Proof of identity and privilege levels need also to apply to services and agents; that way, automated workflows are subject to the same permission models as every other element in the IT stack. The second day of TechEx North America was certainly not a rejection of decision-makers’ AI ambitions – the role of AI and even agents were things of accepted fact among speakers, thought-leaders, and delegates at the event. But there were details and considerations presented by representatives from different industries and business functions, each with positive and insightful things to contribute. Each placed their concerns and their enthusiasms on the table, adding to the discussions around AI implementation in 2026. The march of the robots And there was a great deal of excitement, still, in many areas of the conference floor. The humanoid robots on show were a source of much enthusiasm (everyone seems to love a lovable android!), but more pragmatically, the new Physical AI track drew some of the show’s biggest audiences. Multiple delegates away from the track cited software coding as the place that has first yielded positive results from the use of large language models in professional settings. And from many places too came the opinion that automated physical systems will be the next industry segment set to benefit from concerted work around new models and their practical harnesses. The AI models at the heart of next-gen physical AI are unlikely to be LLMs (although these will be useful if the devices are designed to interact with humans), and as such models develop and emerge from their research stages, it’s the TechEx Events series that will be the first to showcase and present these, and how they can work viably in business contexts. New learning strands to the event This year’s event saw a welcome injection of pragmatic coding, with hands-on learning sessions that took attendees through spinning up their own AI agentic models, with lessons in how agents can improve themselves, right from interactive Google Colab instances. The TechEx Learning Hub also featured workshops from Nvidia and the ever-popular Google Hackathon, with learners ranging in abilities from those that needed introducing to an IDE through to those that came with software skills already well-tuned. Putting learnings into practice is what this event is all about, whether it’s C-suite decision makers taking on lessons on best strategic practices, or developers turning creative ideas into reality. TechEx takes the cutting edge, and distils it through the business lens; pragmatic yet future facing. Catch the next leg of TechEx in Amsterdam this September – who knows how far we may have progressed in the space of four short months? (Image source: TechEx Events) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Enterprise AI roadblocks and roadmaps, security and physical AI: Day two at TechEx appeared first on AI News. View the full article

President Trump flew to Beijing, brought Jensen Huang along at the last minute, and left two days later, telling reporters that “something could happen” on chip exports. Nothing did. Not a single Nvidia H200 has shipped to China since Trump first authorised the sales in December 2025, and US Trade Representative Jamieson Greer told Bloomberg that semiconductor controls were not even on the bilateral agenda. The summit theatre obscured a more interesting development underneath it. The H200 isn’t stuck because Washington won’t allow it. Washington already has. Roughly 10 ******** firms, including Alibaba, Tencent, ByteDance, and JD.com, hold approved US export licences for up to 75,000 units each, with Lenovo and Foxconn authorised as distributors. The chips aren’t moving because Beijing won’t let its own companies take delivery. Two frameworks, one deadlock The mechanics of the stalemate are worth understanding clearly. US rules require that all H200 chips ordered by ******** clients be used only in China. Beijing, meanwhile, has instructed ******** tech companies to limit their use of Nvidia chips to overseas operations while supporting domestic manufacturing. The two requirements are mutually exclusive. Chips cleared for export cannot legally be deployed where Beijing wants to deploy them, and Beijing won’t authorise the domestic use the US licences require, according to Implicator. Commerce Secretary Howard Lutnick stated at a Senate hearing last month that ******** firms are trying to keep their investment focused on domestic suppliers, including Huawei. Beijing’s State Council has also ordered a supply-chain security review aimed at cutting dependence on US semiconductors. The policy contradiction is not accidental. That is the point. What Huawei gained while diplomats talked The days around the summit produced several data points that matter more for the long term than Trump’s parting comment. DeepSeek confirmed its latest model had been optimised to run on Huawei processors. Tencent’s chief strategy officer said ******** GPU supply would increase progressively through 2026, and an Alibaba executive said its T-Head proprietary GPUs had achieved scaled mass production. This follows the April launch of DeepSeek V4, which adapted the model for Huawei’s Ascend chips — the first major ******** frontier model to do so in training, not just inference. What the summit week confirmed is that the shift is no longer experimental. It is now a supply-chain policy. Nvidia’s China revenue has fallen to roughly 5% in recent quarters, down from above 20% before export controls tightened. The company’s own guidance for the current quarter assumes zero revenue from China. Huang’s last-minute inclusion in the delegation–Trump called him directly after seeing media coverage that he had not been invited–suggested urgency. The outcome suggested the limits of what CEO diplomacy can achieve when the obstruction is structural, not procedural. The read for the AI industry The stalemate matters beyond bilateral optics. ******** AI platforms are now operating under a domestic mandate to build on Huawei’s compute stack. The question of which AI hardware architecture becomes dominant in the world’s second-largest AI market is being answered not by technical benchmarks but by government directive. Beijing steering platforms toward Huawei Ascend chips rather than Nvidia H200S is not just a trade posture. It is a structural bet that the performance gap will close fast enough that being locked into the domestic stack is manageable. DeepSeek V4’s results suggest it may be right, at least for inference workloads. Trump said something could happen. Greer said the decision is sovereign for China. Both are true, and neither changes the current position: the H200 deal is approved, licensed, and frozen, with Huawei filling the space it leaves behind. (Image source: The White House) See Also: Can China’s chip stacking strategy really challenge Nvidia’s AI dominance? Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post The Nvidia H200 China deal survived the Trump-Xi summit–just not in the way anyone expected appeared first on AI News. View the full article

Although visitors to an event like TechEx North America will always want to see the cutting edge front and centre stage, the nuance and detail brought to the show by the speakers and exhibitors mean that it’s sometimes the smaller considerations that need to play big – at least, in the minds of enterprise decision-makers. Across the different tracks of Edge Computing, IoT, Data Centre Congress, and Cyber Security, the question was about what needs to be built around AI before it takes its place in the physical, business-oriented world? The Edge Computing track, with its roots in traditional industries, looked at latency, deployment discipline, and cybersecurity for IIoT/IT amalgams. The day-one programme positioned edge computing as a place where companies can reassess the value of their data assets, look at how decisions are made by autonomous equipment, and the required speed of processing. Sessions looked at scaling edge deployments (in multi-site businesses, for example), agentic network operations, distributed inference – on-prem, in-cloud or hybrid – immutable edge infrastructure, and how zero-trust cybersecurity lessons can be applied to control systems. Ed Doran of the Edge AI Foundation chaired a programme that had as its starting point that the edge is a demanding place in which to operate. The track included reps from Akamai, Spectro Cloud, Scylos, TÜV Rheinland, the OPC Foundation, and Germany’s Schneider Electric. Discussions covered issues in manufacturing and IoT, and delved into industrial automation and connected control and attenuation devices. Moving intelligence closer to the machine changes risk profiles (in which direction was a matter for debate), and faster local decisions may reduce latency and dependence on central cloud services, but where do observability and control in decision-makers’ minds? The IoT Tech Expo day-one track on Industrial IoT and Digital Twins looked at manufacturing, with sessions covering smart factory trends, AI beyond Industry 4.0, asset management, practical road-maps for escaping pilot purgatory (more on that below), physical AI in everyday ops, and digital twins. Similar to debates on AI deployment in the knowledge sector, it was the gap between demo and deployment that was the area subject to most scrutiny. Industrial and back office AI both might work well in a presentation, but can stall when they meet old machines (or legacy software). The alliterative pilot purgatory held considerable weight in several sessions on the various presentation stages and on the show floor, day one. The Rockwell Automation and Ford session on physical AI and connected asset intelligence looked especially hard at scaling projects that seem to work well in concept, but may fail when hitting the real world. How does intelligence enter daily operations without becoming another dashboard that nobody owns? Digital twins received similar appraisal. The better version of the digital twin isn’t a visual replica used for demons – although they do have their uses. Instead, several speakers called for, and presented, operational models that can actually help a factory, city, or municipal facility. In addition to pre-testing decisions and improving maintenance, what should the modern digital twin be designed to achieve? The TechEx programme linked ideas between speakers from Siemens, Korea’s LG CNS, Boston Dynamics, and others across the different show strands. The takehome everywhere was that smart systems, be they deeply embedded in engineering sites or the back office, need to be designed in concord with the people or machines that they’re designed to benefit. Day-one sessions at the Data Centre Congress track looked at the big issues facing the sector today: construction, power, procurement, cooling, water, and the network spine needed for AI DCs. Keynote speakers and round-table discussion guests talked about construction chaos and power issues, with the event’s early visitors hearing from TechEx’s host city, Santa Clara, about its own data centre journey. The DC issue remains central to the wider AI debate. As a technology, AI depends on compute, and dense compute at that. This in turn depends on power, cooling, land, and permits. A recurring theme in the infrastructure-focused talks was how AI economics affects the infrastructure stack, with the former rapidly changing, the latter taking years to mature. In many ways, the TechEx event is unique, in that it brings the issues affecting a whole industry under one roof; a place where the ******* picture can be visualised. In the Data Center Congress, we learned that water and power constraints can cut through the rhetoric around the scale of AI. Sessions under the AI and Big Data roof helped temper the idea of a ‘stampede’ to AI productivity, citing their own reasons why unplanned and disorganised implementations of technology don’t fit the modern enterprise. The data centre is now one of the places where AI strategy becomes physical; the enterprise board room’s considerations are practical. The Cyber Security and Cloud Expo track put its own take on deployment forward. Here, the day-one programme dealt with security culture, compliance, speed, ransomware, shadow AI, data exfiltration, legacy systems, open-source dependency issues, and the CISO relationship with the C-suite. There was a general consensus around AI adoption increasing a company’s attack surface, and a much-repeated message that existing security weaknesses don’t diminish when the business wants faster, smarter tools. Sessions on shadow AI and data exfiltration were especially relevant to the wider event. Many companies’ staff use AI services inside business workflows, sometimes without approval, and usually with no facility for logging their activities. That makes data governance and cyber governance effectively the same conversation. The benefits of one conference playing host to complementary tracks were manifest in several cases. For instance, the cybersecurity track’s concerns around legacy systems were echoed on the IoT and Edge stages, where issues were raised about modern, smart intelligence meeting older plant systems. Security in any context can sometimes become an afterthought, but critical infrastructure in the form of transport or energy means that cybersecurity has to play a central role. The TechEx North America day-one tracks that were concerned with infrastructure gave the conference a dose of reality, at least in some respects. AI may be discussed in terms of agentic automation, but deployments depend on networks, data centre capacity, and cybersecurity. Edge and IoT sessions showed how intelligence reaches machines, and how carefully and considerately it needs to be applied. The data centre-focused sessions showed the material limits of physical construction, while the cybersecurity sessions showed how a desire for speed can be the enemy. The day showed the thousands of attendees that putting AI in production isn’t a case of switching the software on. There’s a reliance on the mundane matters of buildings and grids, networks, and security. Companies that understand these issues are more likely to deploy the latest in technology successfully. Getting the ******* picture is what this event is all about. (Image source: TechForge) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post AI is a matter of power, infrastructure and security: TechEx North America appeared first on AI News. View the full article

Amazon has introduced Alexa for Shopping, combining its Rufus shopping chatbot with Alexa+ across its app, website, and Echo Show devices. The assistant can answer product questions, compare items, track prices, and support shopping reminders. It can also handle scheduled shopping actions and eligible automated purchases. The company said Alexa for Shopping combines Rufus’ product expertise with Alexa+’s personalised assistant context. Amazon said Rufus helped more than 300 million customers in 2025 research, compare, and buy products. GeekWire reported that Amazon is retiring the Rufus name from its shopping interface, while Rufus will continue to power parts of the experience behind the scenes. GeekWire also reported that Amazon CEO Andy Jassy said Rufus monthly active users rose more than 115%, while engagement increased nearly 400% year over year. Alexa for Shopping is available through the Amazon Shopping app, Amazon’s website, and Echo Show devices. The feature is rolling out to US customers. Signed-in Amazon customers can use it for free, without a Prime membership, Echo device, or Alexa app. Amazon reported US$426.3 billion in North America net sales and US$161.9 billion in international net sales in 2025. Amazon also reported online stores and third-party seller services as separate revenue categories in its 2025 annual report. Amazon adds shopping questions to search The assistant allows customers to ask shopping-related questions through Amazon’s main search bar instead of using a separate chatbot window. Users can ask for product recommendations or purchase history. They can also ask for advice related to specific shopping needs. Examples shared by Amazon include questions such as “What’s a good skincare routine for men?” and “When did I last order AA batteries?” Amazon said the assistant uses information from its platform to answer these questions. Amazon said Alexa for Shopping uses information from a customer’s Amazon activity and Alexa interactions. That includes shopping history, browsing, purchases, and conversations. Amazon said the information is used to recommend products and support shopping actions. Alexa for Shopping can compare products side by side and provide AI-generated summaries on product pages. It can also show AI-generated overviews in search results with category information. Price tracking and automated shopping Alexa for Shopping can monitor price drops for selected items for up to one year. Customers can view a full year of price history on product detail pages or by asking the assistant. The assistant can create shopping guides for larger purchases. These guides compare product features and prices. They also include reviews from Amazon and the web. Amazon said customers can use the assistant to set scheduled shopping actions, including restocking household items. Amazon said the assistant can also handle birthday reminders and gift suggestions. Scheduled actions can also be tied to conditions. For example, the assistant can add an item to the cart if it reaches a target price and has not been purchased within a set *******. The assistant can search past orders and add frequently purchased items to a customer’s cart through conversational prompts. Amazon said customers can view and update personal details used by Alexa for Shopping. These details can include family members, pets, interests, and dietary needs. Alexa for Shopping can also surface products from other online stores through Shop Direct. For eligible products, Amazon said its Buy for Me agentic AI feature can complete purchases using a customer’s primary address and payment method. Echo Show gets full shopping access Amazon is also adding full-store shopping access to Echo Show. Users can browse, search, and shop using voice, touch, or both. The Echo Show shopping experience is available for Alexa+ customers on Echo Show 15 and Echo Show 21, with support for other devices to follow. Amazon also cited AI investments in its first-quarter 2026 results. The company said free cash flow fell to US$1.2 billion for the trailing 12 months. It attributed the decline mainly to a US$59.3 billion increase in property and equipment purchases, primarily reflecting AI investments. Rajiv Mehta, Amazon’s vice president of conversational shopping, said the assistant can carry customer preferences, past purchases, and conversations across phones, laptops, and Echo devices. Users can access the assistant by updating the Amazon Shopping app and selecting the Alexa icon in the bottom navigation bar. On the desktop, the feature appears at the top of the screen. (Photo by Anirudh) See also: Google tests Remy AI agent for Gemini as focus turns to user control Want to learn more about AI and big data from industry leaders? Check outAI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Amazon launches Alexa for Shopping as Rufus moves behind the scenes appeared first on AI News. View the full article

Enterprise leaders must progress past generative applications and scale “autonomous intelligence” to capture real growth. Generating text or summarising internal communications offers localised productivity improvements, yet these abilities rarely alter the core cost or revenue structure of a large organisation. Enterprises are now focused on deploying systems capable of independent execution. Leaders are demanding applications that can traverse internal networks, execute multi-step logic, and finalise transactions without constant human prompting. Prakul Sharma, principal and AI & Insights Practice Leader at Deloitte Consulting LLP, said: “At Deloitte, we view this as the third stage on an intelligence maturity curve, from ‘assisted intelligence,’ in which AI and analytics help people interpret information, through ‘artificial intelligence,’ with machine learning augmenting human decisions, to ‘autonomous intelligence,’ where AI decides and executes in defined boundaries. “Today’s GenAI-era abilities – like chatbots and conversational AI – sit in the middle of that curve. Agentic AI acts as the bridge into autonomy, and it is where the centre of gravity is changing now. The difference we are seeing is agency: GenAI produces an answer, while autonomous intelligence pursues an outcome by reasoning over a goal, invoking tools and data, and adapting as conditions change, with humans setting guardrails not driving every step. “We’re seeing this show up in industries, and in every case, the unlock isn’t the agent itself, but the surrounding governance architecture of identity and human-in-the-loop checkpoints, making autonomy safe to scale.” Forensic audits for targeted margin improvement To extract actual economic value, these autonomous systems must integrate directly into revenue-generating or cost-heavy workflows. Consider a scenario in enterprise procurement: an agentic application continuously cross-references supply chain inventory against live vendor pricing in an enterprise resource planning system. It can then independently authorise purchase orders in predefined financial parameters, halting only for human approval when deviations occur. The same system must also carry a verifiable identity in the ERP, read pricing data that is current enough to be contractually binding, and operate in approval thresholds that legal and compliance have formally endorsed. Any one of those dependencies, left unresolved, collapses the case for autonomous execution entirely. Achieving this level of automation therefore requires a forensic examination of existing operations before allocating any compute resources. Sharma outlines the method Deloitte uses to initiate this operational overhaul and locate areas where autonomy can generate tangible revenue: “The first step we advise is starting with a decision audit and the process. We ask leaders to pick one or two value chains where outcomes are bottlenecked by decisions not by tasks in that process, and to map how those decisions get made today. We ask questions like who has the data, who has the authority, where the handoffs break, what actions are needed, and where judgement is being applied. “Asking these questions surfaces the process workflows where autonomy will create real economic value, while simultaneously exposing any data and governance gaps that may have derailed a pilot. From there, we help leaders sequence the rewire: stand up the foundational layers with AI and agentic fabric, data, evals, agent identity, and human-in-the-loop patterns against that first value chain, prove it works, and then use it as the template to scale.” Integrating the right data infrastructure and upstream architecture Once the operational target is isolated, the technological execution frequently stalls owing to upstream friction. The underlying foundation models from major providers have advanced quickly enough to handle complex reasoning tasks, becoming largely interchangeable commodities. The friction point lies in connecting these reasoning engines to legacy data architectures. Sharma observes that the true technical barriers emerge long before the prompt reaches the large language model: “Based on what we are seeing, the model is rarely the bottleneck, since frontier ability is now rapidly becoming a commodity. Where enterprises trip up in the design phase is upstream of the model. They select a use case before mapping the underlying workflow, resulting in the agent automating a process that was already broken or poorly instrumented. “The second pattern is data: clients may underestimate that autonomous systems need decision-grade data, not reporting-grade data, meaning lineage and access controls that most enterprise data estates were not built to support.” The distinction matters because most enterprise data estates were built for human analysts, not autonomous systems. Reporting-grade data – aggregated on a nightly or weekly batch cycle, structured for dashboard consumption, and stripped of the lineage that records how a value was derived – is adequate when a person applies judgement before acting on it. An autonomous agent has no such backstop. When it retrieves a contract price or a stock level to execute a transaction, that figure must carry a timestamp current enough to be binding, a traceable provenance, and access controls that confirm the agent is authorised to read and act on it. Providing this decision-grade data involves integrating autonomous agents with right event stores and databases designed to manage both structured and unstructured enterprise information. When an agent retrieves data to execute a task, the enterprise must guarantee its freshness. Relying on stale batch-processed data introduces extreme risk, potentially causing the system to act on obsolete pricing tiers or outdated compliance frameworks. The financial model for scaling these systems also requires forecasting variable compute expenses. Because agentic workflows involve multiple interactions with large language models to reason through a single goal, API costs can escalate unpredictably. Mitigating hallucination risks through retrieval-augmented generation processes also increases the necessary compute overhead, requiring strict financial controls before enterprise-deployment. Reconciling governance debt and enterprise ecosystems Transitioning from controlled testing environments to live enterprise deployment is a very different proposition. A small-scale test might perform perfectly using carefully selected data sets, but deploying that ability in thousands of employees and interconnected software platforms exposes vulnerabilities. Navigating modern enterprise security environments means integrating the agentic architecture deeply with existing identity providers and cloud-native security controls across hybrid cloud ecosystems. Sharma identifies this integration failure and the resulting governance debt that halts progress: “The main roadblock we see is what we call the production gap. A pilot can succeed with a clever prompt, a curated dataset, and a champion team running it manually, but enterprise deployment requires continuous evaluations, identity and authorisation that work in systems the pilot never touched, change management for the users, and a financial model that can absorb use-based costs at scale. “Tied to that is governance debt: the controls, audit trails, and risk frameworks waived to accelerate a pilot often become the gating items once legal and compliance evaluate a production rollout. The clients that break through are ones that don’t treat pilots as experiments but instead treat them as the first production instance of a reusable platform – with the same evals, identity model, and governance. Instead of starting over, this allows the second and third use cases to build on the first.” Compliance frameworks applied during initial testing are often completely insufficient for live deployment. Teams eager to prove a concept frequently bypass standard corporate security protocols, creating the very gating items that prevent future scaling. What unites all three failure modes – the production gap, governance debt, and upstream data friction – is that each one is invisible during a well-run pilot. A champion team with a curated dataset and management cover can paper over missing identity controls, stale data, and deferred compliance reviews for long enough to produce a convincing demonstration. It is only when the system must operate in the full enterprise, with real users, live data, and legal scrutiny, that the gaps become structural blockers not known workarounds. Building a reusable platform from the outset – with identity verification, continuous model evaluations, and financial monitoring treated as first-class requirements not post-launch additions – is what allows organisations to avoid rebuilding those foundations for every subsequent deployment. Prakul Sharma’s interview was conducted ahead of the AI & Big Data Expo North America, where Deloitte is a important sponsor. Be sure to swing by Deloitte’s booth at stand #272 to hear more directly from the organisation’s experts. Prakul Sharma will be sharing more of his insights during a panel session on day one and day two of the industry-leading event. (Image source: Pixabay, under licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Deloitte: Scale ‘autonomous intelligence’ for real growth appeared first on AI News. View the full article

British technology company Humanoid will deploy humanoid robots at factories operated by ******* industrial supplier Schaeffler, Reuters reported. The two companies’ agreement covers an estimated 1,000 to 2,000 robots in Schaeffler’s global manufacturing sites by 2032, according to a Humanoid spokesperson. The companies have not disclosed the contract value. The first deployment is scheduled between December 2026 and June 2027 at two Schaeffler sites in Germany. Humanoid CEO Artem Sokolov told Reuters that the initial phase will cover box handling in Herzogenaurach and near-full-scale factory testing in Schweinfurt. Schaeffler prepares factory rollout Humanoid will also support the integration of the robots into Schaeffler’s existing production lines. The deployment follows a previous technology partnership the two companies announced in January. Schaeffler is testing humanoid robots for physical tasks in production and warehouse environments. The first tasks include moving boxes and handling materials in existing factory setups. Under the supply agreement, Schaeffler will become Humanoid’s preferred supplier for joint actuators through 2031, Sokolov said. The arrangement will cover more than half of Humanoid’s demand for its wheeled humanoid platforms. He added that the agreement is expected to cover at least 1 million actuators over the *******. RLWRLD collects worker motion data The Schaeffler rollout comes as companies in several sectors are becoming test sites for physical AI systems. AP News reported that South Korean AI startup RLWRLD is collecting worker motion data from hotels, logistics sites, and retail settings. At Lotte Hotel Seoul, food and beverage staff have been recorded while folding banquet napkins and preparing tableware. Body cameras placed on the head and hands capture how workers move and grip objects during detailed service tasks. RLWRLD is collecting similar data from logistics workers at CJ, including how they lift and handle goods in warehouses. The company is also working with staff at the Japanese convenience store chain Lawson to track how food displays are organised. It’s building an AI software layer for robots that can operate in factories and other worksites. Its engineers have identified hand dexterity as a priority for industrial and service tasks. Robots learn physical tasks RLWRLD uses human movement data to train its robot systems, converting worker footage into machine-readable data. Engineers then add their own demonstrations using cameras, VR headsets, and motion-tracking gloves. The data captures details like joint angles and the level of force applied during a task, according to Song Hyun-ji from RLWRLD’s robotics team. The company then uses the data to train test robots, including systems guided by human operators wearing control devices. In one demonstration, a wheeled robot with human-like metal hands moved cups at a minibar while guided by engineers. In another, a humanoid opened a box and placed a computer mouse inside. It then closed the box and set it on a conveyor belt. South Korea backs physical AI The company sees industrial deployment as the first target. RLWRLD expects AI robots for industrial use to be deployed at scale around 2028, a timeline it says is shared by some major businesses. Hyundai Motor plans to introduce humanoids built by Boston Dynamics at its global factories in the coming years, starting with its Georgia plant in 2028. Samsung Electronics has said it plans to convert all manufacturing sites into “AI-driven factories” by 2030. Its plans include humanoids and task-specific robots in production lines. Labour groups raise concerns Labour groups in South Korea have raised concerns about the use of worker data and robot deployment. Unions have warned that robot deployment could affect employment and weaken the pipeline for skilled labour. Kim Seok, policy director at the Korean Confederation of Trade Unions, said employers and the government should engage with workers over AI adoption. He said skilled work remains a human achievement. In hospitality, Lotte Hotel is looking at robots for cleaning and other back-of-house work. Current humanoids would need several hours to clean a guest room that human workers complete in about 40 minutes, but the hotel hopes robots will be ready for some cleaning and support tasks by 2029. Lotte Hotel is also considering robot rental services for hospitality and other service industries. Park, one of the hotel workers involved in the training process, said humanoids might eventually take over about 30% to 40% of back-of-house event preparation work. He said tasks involving direct human interaction would remain difficult to replace. (Photo by Andy Kelly) See also: Physical AI raises governance questions for autonomous systems Want to learn more about AI and big data from industry leaders? Check outAI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Physical AI moves closer to factory floors as companies test humanoid robots appeared first on AI News. View the full article

Real estate products depend on integrations, data flows, and compliance layers that rarely appear in marketing pages. A vendor may look strong on reviews about general software development and struggle once MLS feeds, payment systems, and document workflows enter the build. In short, among top real estate app development companies in the US for 2026 are LITSLINK, Code District, Empat, Helpful Insight, and DBB Software. These firms stand out for their work with complex PropTech requirements, including RESO Web API integrations, tenant screening workflows, and full-cycle MVP delivery. Below is a detailed review of their abilities and pricing. Core integrations reliable proptech vendors should master A real estate app is a connector. The core product is often thin. The value sits in what the app talks to. The firms on your shortlist should cover seven categories of integration without asking a discovery question twice. Listing and MLS data. RESO Web API, IDX feeds, and regional MLS providers each use different authentication schemes and field mappings. Identity and credit. TransUnion SmartMove, Experian RentBureau, Checkr, and Plaid Identity each sit behind tenant screening or investor onboarding flows. A vendor that has built a RentBureau pull knows the FCRA disclosures that must sit on top of it. Payments and escrow. Stripe and Plaid handle card and ACH. Dwolla and Modern Treasury handle larger rent and investment flows. Escrow accounts often need a banking partner like Synapse or Treasury Prime. A vendor that has only integrated Stripe Connect is not ready for a rent payment product that touches ACH return codes. Document and e-signature. DocuSign, HelloSign, and Notarize cover lease signing and disclosures. Tax and ownership documents often route through county record APIs where they exist and PDF parsing where they do not. A vendor that can map a multi-state disclosure pack to a document template library saves two months. Mapping and virtual tours. Google Maps, Mapbox, and Here handle base layers. Matterport, Cupix, and Skyword supply 3D tours. A vendor that has wired Matterport SDKs into a mobile app knows why frame rate drops on mid-tier Android phones. CRM and reporting. Salesforce, HubSpot, Follow Up Boss, and kvCORE sit on the agent side. Segment, Amplitude, and Mixpanel cover product analytics. A vendor that has built a two-way Salesforce sync has handled the quirks that one-way webhook integrations never hit. Accounting and finance. QuickBooks Online, Xero, and AppFolio handle the back office for property managers. Fund administration for investment platforms pushes into NAV calculation tools and custody partners. A vendor that calls accounting “out of scope” will push this cost to an in-house team nine months later. In-depth review of top real estate app development companies in the US The companies below demonstrate their abilities through the products they have shipped. Each profile pairs a stated strength with a delivered outcome so you can verify how that strength performs in production. Start with the success story. It shows whether the team handled real estate-specific complexity like listings, transactions, or user workflows. Then match that evidence to your own requirements. CompanyFoundedMin. project sizeTeam sizeKey strengthClient success storyLITSLINK2014$5,000+300+AI-assisted development, 80+ funded startupsRebuilt a 250K+ listing condo marketplace: 20K+ new visitors, 12K+ buildings added, $800K revenue in 3 monthsCode District2017$10,000+250+Award-winning, helped clients raise $10M+Built a real estate task management app on budget and on schedule, leading to a continued partnershipEmpat2013$10,000+250+300+ projects in 17 countries, enterprise clientsDelivered a multi-store delivery platform with UI/UX and QA, praised for design quality and deadline adherenceHelpful Insight2016$5,000+100+2,000+ projects, 92% client retention rateBuilt a convention management app for a real estate developer, still in active use after 6 yearsDBB Software2015$25,000+100+50% faster delivery, 80% of clients stay 7+ yearsBuilt Casavi’s cross-platform React Native app with CI/CD pipelines, instant notifications, and full QA coverage LITSLINK Founded: 2014 Min. project size: $5,000+ Team size: 300+ Provider profile Recognized among the top real estate app development companies in the US, LITSLINK has delivered 300+ products for 200+ clients worldwide, with 80+ of those clients going on to raise venture follow-on funding. Their 300+ in-house team covers architecture and design under one roof, with daily commits, sprint-end demos, and a dedicated on-call manager around the clock. LITSLINK applies AI-assisted development in the full build cycle to cut delivery time and reduce cost without affecting output quality. For engagement structure, clients choose between fixed-price projects, monthly staff augmentation, or full-cycle outsourcing. Client success story A condo marketplace with 250,000+ listings needed a full product overhaul. LITSLINK rebuilt the brand identity, replaced the front end with React.js, and developed a custom targeting tool designed specifically for agents. Three months after launch, the platform had attracted 20,000+ new visitors, added 12,000+ buildings to its database, and generated up to $800K in revenue. Code District Founded: 2017 Min. project size: $10,000+ Team size: 250+ Provider profile Code District is an award-winning digital transformation company with 200+ developers and DevOps engineers on staff. They have been rated the most recommended software development company three years running (2020, 2021, 2022) and hold top rankings for app modernization and AI development. They run on agile methods, offer project-based and dedicated team models, and have helped clients cut cloud costs from $30K to $10K and raise over $10M in funding in two separate engagements. Client success story A real estate client needed a mobile app built around task management, status tracking, and calendar functionality. Code District handled the database architecture and frontend development, keeping the user experience clean and intuitive throughout. The project came in on budget despite a demanding timeline. Empat Founded: 2013 Min. project size: $10,000+ Team size: 250+ Provider profile Empat delivers full-cycle software development in mobile and system architecture. They have completed 300+ projects in 17 countries for clients including Porsche, Panasonic, CBRE, and Transparency International. Their 140+ in-house experts cover a technical stack: React, Vue, Angular, Flutter, React Native, Node.js, Python, Ruby on Rails, Go, and more, supported by solid DevOps and QA practices. PropTech sits among their named verticals. The company offers fixed price, dedicated team, and time and material engagement models. Client success story Empat designed and developed a website for a multi-store delivery platform, handling UI/UX design and QA testing with the build. Stakeholders in the organization pointed to the user-friendly design as a standout result. The team managed tasks through Jira, kept to deadlines, and brought proactive ideas to the project throughout the engagement. Helpful Insight Private Limited Founded: 2016 Min. project size: $5,000+ Team size: 100+ Provider profile Helpful Insight is a mobile and AI development company with 10 years of experience, 2,000+ projects delivered, and a 92% client retention rate. They serve clients in 40+ countries and 30+ industries, with real estate among their dedicated practice areas. Their real estate abilities cover property management systems, tenant management tools, online booking platforms, auction portals, IDX integration, lead management apps, and investment platforms. The company offers custom builds, dedicated teams, and staff augmentation. Client success stories A real estate development company needed a mobile app to manage their convention operations. Helpful Insight built a custom solution that combined property listings, consultation scheduling, and document ordering into a single platform. The app made on-site event management more organized and efficient for property stakeholders. The client has continued using the solution for six years, citing consistent delivery and a creative approach to ongoing improvements. DBB Software Founded: 2015 Min. project size: $25,000+ Team size: 100+ Provider profile DBB Software is a certified AWS Partner with 10+ years of experience, 100+ engineers, and a client lifetime average of six years. 80% of clients stay for 7 or more years. Their structural advantage is speed: a library of pre-built solution blocks cuts development time by 50%, reduces POC and prototyping stages by 30%, and halves planning time. In real estate, they cover custom platform development, mobile apps, UI/UX, system integrations, architecture reviews, and ongoing maintenance. Client success stories DBB Software built a cross-platform React Native app for Casavi, covering both iOS and Android from a single codebase. The team added instant notifications, in-app damage reporting, and a news and updates section. They configured automated CI/CD pipelines via Fastlane to speed up branded version releases in Casavi’s client base, and covered the full QA cycle with Jest and Detox testing. A 5-phase implementation roadmap for real estate apps A real estate app that reaches production in 9 months follows a 5-phase plan. Skip a phase, and the cost shows up in month seven. Phase 1 – Discovery and integration mapping (weeks 1 to 4). The output includes a signed scope, a wireframe set for the priority flows, and a named list of every external system the app will interact with. The integration list is the deliverable that protects the budget. Phase 2 – Architecture and compliance review (weeks 4 to 7). Cloud topology, data residency, encryption at rest and in transit, and the audit trail design sit in this phase. Fair Housing, CCPA, GLBA, and state-specific rules apply depending on the product. A compliance checklist signed by the client’s legal team is the end of the phase. Phase 3 – MVP build (weeks 7 to 20). Two-week sprints, one demo every two weeks, and one integration hook wired per sprint. The priority order is identity first, listings second, payments third, and notifications fourth. Phase 4 – Beta with twenty real users (weeks 20 to 26). Ten agents or property managers, ten renters or buyers, six weeks of live use in one metro area. The metrics are listing-to-lead rate, time-to-application, and payment success rate. Every failure mode that appears in beta is cheaper to fix than the same failure in month twelve. Phase 5 – Production and a 90-day hypercare window (weeks 26 to 38). Staged rollout by region, uptime and latency SLOs in writing, an on-call rotation with named engineers, and a weekly defect review. The 90-day window closes with a support handoff document that lists every runbook, every alert threshold, and every integration credential owner. Final thoughts Vendor selection shapes the next two to five years of your product. The wrong choice slows delivery, increases maintenance costs, and limits how far the platform can scale. To make the right choice, focus on three signals. First, integration experience in MLS, payments and documents. Second, proof of delivery through real products in use. Third, a clear roadmap that defines how the product moves from idea to production. Shortlist two or three vendors that match your scope and budget. Validate their recent real estate work, confirm their integration history, and review their delivery plan in detail. The post Top real estate app development companies in the US: Abilities and costs appeared first on AI News. View the full article

The Physical AI Conference shaping the future of robotics, autonomous systems and real-world AI deployment lands in Silicon Valley this May, bringing together the engineers, builders and AI pioneers turning intelligence into physical action. Physical AI Expo North America will take place on May 18–19, 2026 at the San Jose McEnery Convention Center, uniting global AI innovators, robotics leaders, enterprise technologists and next-generation infrastructure providers for two days focused on the future of AI in the physical world. As AI rapidly moves beyond chatbots and software into robotics, industrial automation, autonomous systems and intelligent machines, the race is now on to operationalise Physical AI at scale. From manufacturing and logistics to automotive and defense, organizations are investing heavily in AI systems capable of sensing, reasoning and acting in real-world environments. Physical AI Expo has been created to explore exactly how this transformation is happening – and what it takes to move from experimentation to production. The Rise of Physical AI The next era of AI is physical. While software-based AI has transformed digital workflows, the next competitive leap comes from embedding intelligence into machines, robotics, autonomous systems and industrial operations. Physical AI Expo will examine how organisations are deploying AI-powered systems into real-world environments, integrating intelligence into operations, and building infrastructure for AI that interacts with the physical world. The conference agenda combines deep technical insight with enterprise-scale deployment strategies, bringing together leaders developing the next generation of intelligent systems. Topics include: Physical AI & Robotics at Enterprise Scale Autonomous Systems & Industrial Automation AI Infrastructure, Data Platforms & Compute Enterprise AI Deployment & ROI Human-AI Collaboration AI Reliability, Transparency & Safety From Prototype to Production AI Developer Tools & Workflows From Prototype to Production Across two days, attendees will hear how global technology leaders are deploying Physical AI in production environments and solving the real challenges around scalability, infrastructure, reliability and adoption. Day One: AI Strategy, Infrastructure & Data Day One focuses on AI strategy, enterprise transformation, autonomous intelligence and large-scale data infrastructure. Sessions are designed for organizations building the foundations required to deploy Physical AI systems at scale. Day Two: Robotics, Autonomous Systems & Real-World AI Day Two explores how companies are moving Physical AI from prototype to production across robotics, automation and autonomous operations. Dedicated tracks will cover robotics, AI deployment, infrastructure and developer workflows powering intelligent systems in the real world. Physical AI Leaders Taking the Stage The Physical AI Conference program features experts from some of the world’s most advanced AI and robotics organizations, including: Leslie Karpas, Inception Global Head of Physical AI, NVIDIA Arne Stoschek, VP AI and Autonomous, Airbus Acubed Jose Alvarez, Director of Research, NVIDIA Dr Vinesh Sukumar, Vice President of AI, Qualcomm Simon Ninan, SVP & Global Head of Strategy, Hitachi Sungho Kim, CEO, Hyundai Global Software Center Naresh Dulam, Senior VP of Software Engineering, JPMorgan Pierre-Alexandre Balland, Chief Data Scientist, CEPS & Co-founder, General Robotics The event brings together leaders across robotics, AI infrastructure, industrial automation, automotive, logistics and intelligent systems engineering – creating one of the most important Physical AI gatherings in North America. Building AI That Acts in the Real World Michael Hughes, Head of Conference Production, commented: “Physical AI is rapidly moving from concept to deployment. The conversation is no longer just about models – it’s about infrastructure, robotics, autonomous systems and building AI that can reliably operate in the real world at scale. Physical AI Expo brings together the companies and engineers making that future happen right now.” About Physical AI Expo North America Physical AI Expo North America is a leading Physical AI Conference focused on robotics, autonomous systems, AI infrastructure and enterprise deployment. The event connects technology leaders, engineers, innovators and developers exploring how AI is transforming physical operations and intelligent systems worldwide. Physical AI Expo North America takes place May 18–19, 2026 at the San Jose McEnery Convention Center. The post Physical AI Conference Comes to San Jose as Robotics & Autonomous AI Go Mainstream appeared first on AI News. View the full article

Joe Rose, president at strategic technology provider JBS Dev, wants to cut through one of the myths of working with generative and agentic AI systems. “It’s a common misconception that your data has to be perfect before you do any of these types of workloads,” he explains. As a recent article in AI Fieldbook outlines, vendors and consultants – not surprisingly – suggest you need huge data lakes and multi-year data transformation programmes respectively. Executives are therefore scratching their heads at it all. The reality is slightly different. “The tooling has never been better than it is now to deal with poor quality data,” says Rose. “It’s almost remarkable what an LLM can understand on a half-written prompt.” It makes sense. If you’ve got such a tool available, then it’s worth utilising that to your advantage – with the correct guardrails in place. The inherent unpredictability of models means a need to handle bad output, which is where the human in the loop comes in. For textual or category data, there is a resilience in place. “People are… used to ‘we build it, it works, we forget about it,’” says Rose. “That’s just not how these systems work.” Regarding imperfect data, Rose gives an example of a client in the medical sector where the goal was to migrate to another billing reconciliation system. Records were a mix; some were in PDF, others an image; the procedure would sometimes be in the doctor’s name, the doctor’s name would be in the patient’s name, and so on. The gen AI was able to scope the clean data from a simple prompt, from OCR to the images to text extraction for the PDFs, while more agentic approaches were subsequently leveraged, such as comparing a customer record to an insurance contract to see if they were billed at the right rate. “You start to layer different use cases on top of one another,” says Rose. “That’s not to say that it gets everything right – you still need a human in the loop. But what you want to do is say, ‘we started at 20% automated, and then 40%, and then 60, 80%’, and kind of grow that over time.” Going forward, Rose expects future discussions for these models to be around cost and portability. “I think you’re going to see a shift away from these radical leaps and model capability, and more shift towards ‘how do we make the cost more sustainable that we don’t have to build data centres at the rate we’re building data centres?’,” he says. “The last mile is ‘how do we get these things to run on a laptop or a phone instead of having to run in a data centre?’ The models were trained on a body of data – essentially every page on the internet and other stuff. It’s not like there’s a tonne more data that hasn’t already been put into them that’s going to lead to some type of breakthrough.” At AI & Big Data Expo, where JBS Dev is participating, Rose is looking forward to the conversations – and one more controversial opinion he’ll put across is to tell folk to stop buying from SaaS vendors when you can do it yourself. “It’s not as hard as it sounds,” he says. “Almost everybody’s got some kind of cloud presence, and that’s where I would start, because the cloud tooling, especially for the big three… has everything you need to start implementing agentic workloads tomorrow, without new software licenses and new training.” Once that’s in place, JBS Dev is there for the next steps of the journey. Watch the full interview with Rose below: Image by Gerd Altmann from Pixabay The post JBS Dev: On imperfect data and the AI last mile – from model capability to cost sustainability appeared first on AI News. View the full article

A malicious Hugging Face repository that posed as an OpenAI release delivered infostealer malware to Windows machines and recorded about 244,000 downloads before removal, according to research from AI security firm HiddenLayer. The number of downloads may have been artificially inflated by the attackers to make the model seem more popular, so the extent of the effects of the attack is unknown. ‘Open-OSS/privacy-filter’ imitated OpenAI’s Privacy Filter release. HiddenLayer said the original model card had been copied nearly exactly, and the bad actors included a malicious loader.py file that fetched and ran credential-stealing malware on Windows hosts. The repos reached the top of the ‘trending’ list on Hugging Face with 667 likes accrued in less than 18 hours – again, this figure may have been changed by the attackers. Public AI model registries may be becoming risks in the software supply chain as developers and data scientists clone models directly into corporate environments, environments that have access to source code, cloud credentials, and internal systems. That situation alone makes a compromised model repository more than a nuisance. The README file for the fake model closely resembled that of the legitimate project, but it departed from the original in that it instructed users to run start.bat on Windows or execute python loader.py on Linux and macOS, instructions central to the infection chain HiddenLayer described. Researchers have previously warned that malicious code can be hidden inside AI model files or related setup scripts on Hugging Face and other public registries. Previous cases involved Pickle-serialised model files that bypassed platform scanners. Malicious loader disguised as setup code HiddenLayer said loader.py began with decoy code that resembled a normal AI model loader, moving quickly to a concealed infection chain. A script disabled SSL verification, decoded a base64-encoded URL linked to jsonkeeper.com, retrieved a remote payload instruction, and passed commands to PowerShell on Windows machines. HiddenLayer said the use of the command-and-control channel jsonkeeper.com allowed the attacker to rotate the payload without changing the repo’s contents. The PowerShell command then downloaded an additional batch file from an attacker-controlled domain, and the malware established persistence by creating a scheduled task designed to resemble a legitimate Microsoft Edge update process. The final payload was a Rust-based infostealer. According to HiddenLayer, it targeted Chromium and Firefox-derived browsers, Discord local storage, cryptocurrency wallets, FileZilla configurations, and host system information. The malware also tried to disable Windows Antimalware Scan Interface and Event Tracing. Wider campaigns HiddenLayer also said it found six further Hugging Face repositories containing virtually identical loader logic that shared infrastructure with the cited attack. The case follows other warnings about malicious AI models on Hugging Face, including poisoned AI SDKs and fake OpenClaw installers. The common thread is that attackers are treating AI development workflows as a route into normally secure environments. AI repositories often contain executable code, setup instructions, dependency files, notebooks, and scripts, and its these peripheral elements that cause the problems, rather than the models themselves. Sakshi Grover, senior research manager for cybersecurity services at IDC, said traditional SCA was designed to inspect dependency manifests, libraries, and container images. It is less effective at identifying malicious loader logic in AI repositories. They also cited IDC’s November 2025 FutureScape report, which contained the call that by 2027, 60% of agentic AI systems should have a bill of materials. This would help companies track which AI artefacts they use, their source, which versions were approved, and whether they contain executable components. Response and mitigation HiddenLayer advised anyone who cloned Open-OSS/privacy-filter and ran start.bat, python loader.py or any file from the repository on a Windows host to treat the system as compromised, and recommends re-imaging systems. Browser sessions should considered compromised even if passwords are not held locally, as session cookies let attackers bypass MFA in some circumstances. Hugging Face has confirmed the repo has been removed. (Image source: Pixabay, under licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Hugging Face hosted malicious software masquerading as OpenAI release appeared first on AI News. View the full article

Laserfiche has announced the release of AI agents that can help perform tasks through natural language prompts. Intelligent assistants follow Laserfiche’s integrated security rules and compliance requirements, helping ensure all sensitive data remains protected. Karl Chan, CEO of Laserfiche, said, “The introduction of AI Agents to content management signals a change in how we handle the information lifecycle. We are moving beyond manual processes by offloading mundane work to agents that operate in a governance framework. We are letting organisations modernise operations while keeping compliance at the forefront.” Laserfiche’s AI agents use generative LLM reasoning models that perform actions, potentially cutting time resource spend by handling the middle ground between the design of automated workflows and manual tasks. Through document data analysis, the agents can operate tasks and make changes based on natural language user instructions. Laserfiche AI agents abilities Laserfiche agents are accessed via Smart Chat, a chat interface, with what agents are able to perform limited to the user’s permissions and restrictions. This ensures teams and users of different technical levels can use the tools to automate their work more safely. Through a blend of intelligent agents and AI-driven content analysis, organisations can identify specific information in documents, letting them take steps in departments such as legal, accounts payable, and HR. In legal circles, Laserfiche AI agents can spot inconsistencies in documents and contracts before routing them for human review. Accounts Payable can use the agents to find late invoices and direct them to the necessary teams to be resolved. In HR, the AI system can scan employee records (age, gender, address, for example) and identify details that will move certain documents to the correct digital folders, based on the user’s security level. Agents in industry Laserfiche AI agents have been designed to filter content from repositories and make context-aware action, helping users search for and organise information. Justin Pava, Laserfiche chief product evangelist, spoke on the future of document storage, saying “the ‘where’ of document storage is not going to be as important as it used to be. With automatically-extracted metadata, AI-assisted search and the autonomous abilities of Laserfiche AI agents, you won’t have to spend time organising data, you will be able to simply act on it.” Available for users of Laserfiche Cloud from May 7, 2026, users can direct the company’s AI agents to perform “one-time actions from […] Smart Chat.” Further updates will enhance the agent’s abilities, like embedding them into business processes, letting agents run in the background, and monitor systems for certain conditions. (Image source: Pixabay, under licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Laserfiche unveils AI agents for natural language workflows appeared first on AI News. View the full article

Artificial intelligence is transforming how companies handle compliance. Background checks run in real-time. Payroll monitoring flags discrepancies automatically. Predictive analytics anticipate employee churn before it happens. HR tech stacks now offer automated solutions for nearly every regulatory requirement – from GDPR data requests to workplace safety reporting. But there is one glaring exception. For *** tech companies whose competitive advantage depends on hiring international AI talent, the compliance function that matters most remains stubbornly analogue: sponsor licence management. This creates a dangerous paradox. The sector building the most sophisticated automation tools cannot automate its own immigration compliance. And the consequences are not theoretical. They are immediate and increasingly common – for both employers and the skilled workers who depend on them. The irony tech founders don’t see coming Walk into any London tech scaleup and you will find teams building compliance automation. One might be developing AI-powered contract review. Another could be creating real-time financial reporting dashboards. A third might be launching automated cybersecurity monitoring. These same companies then handle their sponsor licence obligations using spreadsheets, email reminders, and institutional memory. The gap is striking – and it stems from a structural reality most founders do not anticipate. The Home Office Sponsor Management System was not designed for API integration. Compliance data lives in PDFs and manual entries, not structured databases. Material changes to sponsored workers’ circumstances – the kind of events that trigger reporting obligations – require human judgement to identify and interpret. When a machine learning engineer’s role evolves from individual contributor to team lead, no algorithm flags that this constitutes a “material change in job duties” requiring notification in 10 working days. The result: tech companies accustomed to automating risk out of their operations are managing sponsor compliance the same way businesses did in 2010. Manually. Inconsistently. And often incorrectly. For a sector where 30% to 40% of the workforce holds Skilled Worker visas, this is not a minor process inefficiency. It is a systemic operational risk sitting in the least automated corner of the business. The real stakes for *** tech – and the workers caught in the middle The numbers tell the story clearly. Between July 2024 and June 2025, 1,948 sponsor licences were revoked in the *** – more than double the previous year. Analysis of Home Office enforcement data shows the tech sector is disproportionately represented in these revocations, not because tech companies are more reckless, but because they are structurally more vulnerable. AI and machine learning roles are among the hardest to fill domestically. The talent pipeline for specialists in natural language processing, computer vision, and reinforcement learning remains heavily international. A Cambridge-based AI startup competing for Series B funding cannot wait six months to fill a senior ML engineer role with a domestic candidate who may not exist. They hire the best person globally and sponsor them. This dependency creates exposure. When a sponsor licence is suspended, all sponsored workers’ visas are curtailed to 60 days. For a scaleup with 15 AI engineers on Skilled Worker visas, that is not a staffing adjustment – it is an existential threat to product timelines, investor confidence, and competitive positioning. But the human cost runs deeper. A skilled worker who relocated their family to the ***, enrolled children in schools, signed a two-year lease – they suddenly have 60 days to secure a new sponsor or leave the country. Their career trajectory, their children’s education, their financial stability all hinge on finding an employer willing to transfer sponsorship in a two-month window. The financial impact extends beyond direct replacement costs. One mid-sized London fintech lost its licence after a compliance visit uncovered unreported changes in multiple sponsored workers. Eight engineers left in the 60-day window. Three went to competitors. Two returned home. The company faced a 12-month prohibition on applying for a new licence. Eighteen months later, they still had not fully rebuilt their machine learning team. The Series B round they were planning never materialised. “The businesses facing enforcement action are rarely the ones cutting corners deliberately,” says Yash Dubal, director at A Y & J Solicitors, which advises on Skilled Worker Visa applications and compliance. “They are organisations that built a workforce carefully, sponsored overseas workers through the proper channels, and then – somewhere in the day-to-day pressure of running a business – allowed the ongoing compliance framework to drift.” At A Y & J Solicitors, which helps professionals and businesses navigate the Skilled Worker Visa route, this pattern emerges repeatedly. Tech companies treat immigration compliance as an HR administrative task not what it actually is: a business-critical governance function sitting at the intersection of talent strategy, regulatory risk, and operational continuity. The irony is that the solution requires exactly the kind of thinking tech companies excel at – just applied to an unfamiliar domain. What tech founders consistently miss The failure mode is predictable. It starts with assumptions that do not hold. Assumption one: Compliance is like other HR functions. It is not. Payroll errors can be corrected. Missed performance reviews have no regulatory consequence. Sponsor licence breaches trigger enforcement action. There is no grace *******, no software patch, no “we’ll fix it in the next sprint.” The Home Office does not operate on agile principles. Assumption two: There must be a software solution. There is not. The market has produced sophisticated tools for nearly every other compliance challenge, but sponsor licence management remains resistant to full automation because the Home Office systems themselves are not built for it. The regulatory framework pre-dates API-first architecture by decades. Assumption three: Complexity is overstated. It is not. A material change in a sponsored worker’s circumstances must be reported in 10 working days. What constitutes “material”? A salary increase that pushes total compensation above the original Certificate of Sponsorship amount. A change in job title. A change in working location. A change in working pattern that alters the nature of the role. All of these require human judgement to identify in real-time in a fast-moving organisation. Assumption four: Our people know what to do. They do not – not without systems. When an AI engineer gets promoted to lead a team, does the engineering manager know this triggers a reporting obligation? Does the HR business partner? Does payroll? In most tech companies, the answer is no. The knowledge exists somewhere, usually in the head of one person who joined three years ago and remembers the licence application process. That is not a system. It is a single point of failure. “I have sat with clients who believed they were fully compliant, received an inspection, and discovered that what they thought was minor administrative imprecision was, in the Home Office’s view, a pattern of systemic non-compliance,” Dubal explains. “The gap between those two interpretations is where licences are lost – and where skilled workers’ lives are upended.” The companies that navigate sponsor compliance successfully are not necessarily better resourced. What differentiates them is that they have applied engineering discipline to a legal obligation. They have built systems. The systems thinking solution Treating sponsor compliance like an engineering problem changes how it gets managed. First, define the system boundaries. What events trigger reporting obligations? Job title changes. Salary adjustments above thresholds. Role responsibility shifts. Working location changes. Absences exceeding defined periods. Each is a signal that must be captured and acted on. Second, create forcing functions. In software development, automated tests prevent broken code from reaching production. The sponsor compliance equivalent is integrating checks into existing workflows. When HR processes a promotion, the system prompts: “Does this person hold a Skilled Worker visa? If yes, review reporting obligations.” When payroll processes a salary increase, the same check occurs. The compliance step is embedded, not optional. Third, establish verification loops. Quarterly internal audits replicating what a Home Office inspector would examine. Payroll records cross-referenced against Sponsor Management System entries. Employment contracts checked against actual job duties. The gaps surface before an inspector finds them. Fourth, assign clear ownership. In tech companies, product quality has an owner. Security has an owner. Sponsor licence compliance needs the same governance structure – a named individual with authority and board visibility. Not as an add-on to someone’s existing role, but as a function with defined responsibility. Fifth, document everything. If the process for reporting a material change exists only in one person’s understanding of “how we do things,” it will fail the moment that person is unavailable. Documentation creates institutional resilience. It allows the process to work the same way regardless of who is executing it. This is not revolutionary thinking for tech companies. It is how they already manage code deployments, infrastructure changes, and data governance. The challenge is recognising that sponsor compliance deserves the same operational rigour. The questions every tech board should ask The paradox remains: the sector most capable of building automated compliance systems cannot yet automate its most critical compliance function. But tech founders are problem solvers. The path forward requires asking three questions: Redundancy: If our Head of HR left tomorrow, does the step-by-step process for a “Change of Circumstance” report exist in a shared manual, or is it in their head? Integration: Is our immigration lawyer a firefighter we call when things go wrong, or are they an architect helping us build these internal checks? Visibility: Does the Board understand that a simple 11-day delay in reporting a salary bump could technically trigger a 60-day countdown for 40% of our engineering staff? The answers reveal whether sponsor compliance is treated as a system or as tribal knowledge. In a sector built on eliminating single points of failure, that distinction matters – not for the business, but for every skilled worker whose *** future depends on getting it right. The post AI automates HR compliance, except for the area tech companies need appeared first on AI News. View the full article

Bain & Company has estimated a US$100 billion market in the US for SaaS companies using agentic AI. The firm said the market is tied to automating coordination work in enterprise systems. The estimate comes from the second report in Bain’s five-part series on the software industry in the age of AI. The report examines where agentic AI could create new software markets and how SaaS companies can capture them. Coordination work in enterprise systems Bain said the market lies in the manual work employees perform between enterprise applications. These workflows often span ERP, CRM and support systems. They may also involve vendor management tools and email. That work includes pulling data from one system and checking it against another source. It can also involve interpreting unstructured messages and deciding whether to approve, respond, escalate, or wait. Bain said rules-based automation and robotic process automation are limited in workflows involving ambiguity and information spread in multiple systems. Agentic AI can interpret information from different sources, coordinate actions in systems, and operate in policy guardrails. The report argues that agentic AI is not primarily a replacement for SaaS platforms, but that the market comes from converting labour-intensive coordination work into software spending. It estimates vendors are already capturing US$4 billion to US$6 billion of the US market. More than 90% remains untapped, according to the firm. Outside the US, Bain estimated that Canada, Europe, Australia, and New Zealand could add a similar-sized market. That would bring the total in those regions and the US to about US$200 billion. Market size by function The market is not evenly distributed in enterprise functions. Bain estimates that sales represents the largest single share at about US$20 billion. This is mainly due to the number of sales employees, not unusually high automation potential. Cost of goods sold and operations account for about US$26 billion. The large size of the operational workforce means even modest automation rates can translate into a large addressable market. R&D and engineering, customer support, and finance each represent about US$6 billion to US$12 billion in addressable market size. These functions have sizeable workforces and higher automation potential in specific workflows. Customer support and R&D or engineering have the highest automation potential, with roughly 40% to 60% of workflow tasks automatable. Bain said both areas have structured data, standardised processes, and clearer output signals. Finance and human resources fall in the 35% to 45% range. The report said accounts payable and payroll have higher automation potential, while financial planning and employee relations involve more judgement. Sales and IT sit at 30% to 40%. Bain pointed to relationship nuance, deal-by-deal variation, and the unpredictable nature of security incidents as limits on automation in those areas. Legal has lower overall automation potential, at 20% to 30%. Bain said contract review and compliance are repeatable, but the consequences of errors create a need for tighter oversight. Bain’s automation factors The report identifies six factors that determine how much of a workflow can realistically be handled by an AI agent. They include output verifiability, consequence of failure, digitised knowledge availability, and process variability. Bain said workflows with clear verification signals are easier to automate than work involving subjective judgement. Examples include compiling code, reconciled invoices, and resolved support tickets. Workflows involving regulatory or financial risk require closer human supervision, even where agents are technically capable, according to the report. These include tax filings, legal compliance, and security incident response. Bain also identified digitised knowledge availability as a constraint. Agents need access to structured data and documented context. They also need machine-readable inputs, including decision logic that often sits informally with experienced employees. Integration complexity affects automation when workflows pass through several systems and APIs. Authentication layers and exception-handling processes add further complexity, and these workflows are harder to automate end-to-end than workflows contained in a single platform. The highest-value areas are concentrated where no single system of record controls the full outcome. These workflows often span ERP, CRM and support systems, the company says. David Crawford, chairman of Bain’s global technology and telecommunications practice, said SaaS companies have spent the past two decades building positions around systems of record with the next source of advantage being “cross-workflow decision context,” which is defined as the ability to interpret and act in workflows that move through multiple systems. Company examples and adjacent workflows The report cited Cursor, Sierra, Harvey, Glean, Salesforce, ServiceNow, and Workday in its discussion of agentic AI adoption. Cursor has surpassed US$16.7 million in average monthly revenue, according to Bain, after doubling in a single quarter. Sierra has crossed US$150 million per annum, Harvey passed US$190 million pa, and Glean US$200 million pa. The report also pointed to GitHub for example of a company using data from an existing core workflow to move into adjacent work. GitHub’s core business is developer collaboration and source control, but its repository and workflow data helped support expansion into AI-assisted developer productivity and security automation. Bain said SaaS companies can expand through two types of workflow automation. The first is automating core workflows, where they already have domain knowledge and customer trust. Bain said existing system integrations can support automation of core workflows. The second is automating adjacent workflows that the company does not currently serve directly. These areas can be harder to identify because they require detailed mapping of customer workflows and the underlying data that supports decisions. Pricing models can change when agents deliver completed outcomes. Bain said outcome- and use-based pricing can become more relevant when agents resolve issues or process invoices. The report contrasts this with traditional pricing based on seats and logins. Bain’s recommendations for SaaS companies Bain recommended that SaaS companies begin by identifying which customer workflows are now automatable with agentic AI. The firm said companies should assess automation at the subprocess level not treating entire functions as equally automatable. The report also said companies should assess the quality of their data. Bain said relevant factors include whether the data is comprehensive, tied to outcomes, and usable for automation. Bain said companies could close ability gaps through internal development, acquisitions, or partnerships. The report cited AppLovin’s in-house development of its Axon platform, ServiceNow’s acquisition of Moveworks, and Salesforce’s partnership with Workday as examples of different approaches. The firm also pointed to the need for AI engineering talent, cloud-native architecture for multi-agent orchestration, and funding for model training and inference. It said companies should align pricing and sales incentives with AI-driven outcomes not legacy seat-based models. Bain said SaaS companies will also need data and product foundations designed for agentic workflows, including machine-readable hand-offs and systems that capture decisions and outcomes from each workflow run. Crawford said the timeframe for SaaS companies is “measured in quarters, not years,” as AI-native companies gather more deployment data with each customer workflow they automate. (Photo by engin akyurt) See also: Google tests Remy AI agent for Gemini as focus turns to user control Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Bain sees US$100 billion SaaS market in agentic AI automation appeared first on AI News. View the full article

RingCentral has expanded its AI Receptionist product with new links to Shopify, Calendly and WhatsApp, as the communications software company tries to push the product beyond basic call answering and into more routine customer service tasks. The company said AI Receptionist, known as AIR, can now handle some order enquiries through Shopify, arrange appointments through Calendly, and respond to inbound WhatsApp messages. AIR is also being added to shared SMS inboxes and call queues, so it can answer texts and step in when phone lines are busy or staff are not available. RingCentral said more than 11,800 businesses now use AIR. The product is aimed mainly at smaller and mid-sized organisations that receive regular inbound enquiries, and RingCentral cited healthcare, financial services, legal, hospitality, and construction as areas where customers are using AIR for front-desk tasks and after-hours cover. Keller Interiors, an installation company working for Lowe’s Home Improvement, said it deployed AIR in 33 locations. Beth Owens, chief of staff, said the company had a routing problem that was difficult to solve with staff. “RingCentral AIR solved a problem we didn’t have a good human answer for, how do you route every inbound call correctly, 24/7, across 33 locations, without building a call centre?” Owens said. She said Keller Interiors had reduced waiting times from 12 minutes to 90 seconds and saw customer satisfaction scores rise by three points in the course of four months. Tara Breaux, vice-president of operations at Maple Federal Credit Union, said it used AIR to reduce hold times in branches. “We’ve reduced hold times by 90%, enabling faster service, less strain on staff, and more focus on the conversations that matter most.” The new Shopify link is designed to let AIR answer basic questions about orders and customer support over the phone. The Calendly interface lets AIR schedule appointments using tools from Calendly, and using WhatsApp extends into the messaging app used widely by consumers and small businesses. RingCentral is also adding automatic language detection. The company said AIR can recognise a caller’s language and continue the conversation in that language, offering 10 languages, including English, Spanish, French, Italian, *******, and Portuguese. Michelle Morgan, research manager for AI-enabled sales, customer service and contact centre strategies at IDC, said the update was an example of applied AI in daily business. “RingCentral’s expansion of AIR into Shopify, Calendly, WhatsApp, and intelligent call queues shows what applied AI should look like: every feature tied to a clear pain point,” she said. Joe Fahrner, RingCentral’s vice-president of growth for AI products, gave the company’s more expansive view of the product, saying AIR is becoming a “digital employee” for small and mid-market businesses. RingCentral said AIR is now available as a standalone product starting at $49 a month, including 100 minutes. Existing RingEX customers can add AIR starting at $39 a month, also including 100 minutes. (Image source: Pixabay, under .) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post RingCentral adds Shopify, Calendly, and WhatsApp to AI Receptionist appeared first on AI News. View the full article

The words “pressure” and “NHS” go hand in hand in the *** and unfortunately there is no sign of a reduction in the strain the institution suffers any time soon. As NHS England continues the struggle to reduce its 7.25 million waiting list, new policies are being introduced to move care away from hospitals and into the community, despite GPs’ warning of increased workloads and risk to patients. Add in looming doctor strikes and deepening staff shortages and the backdrop of the health service does not look rosy. In a bid to relieve some of the burden, AI-enabled virtual care is emerging as a tool to manage the growing number of patients outside hospital settings. The technology is being implemented to help around three important areas – waiting lists, hospital capacity, and corridor care. Michael Macdonnell, Deputy CEO at European virtual care provider Doccla, who has first-hand experience working in the NHS, commented, “The NHS is facing unprecedented pressure, with a 7.2 million patient waiting list, patients waiting in ambulances and in corridors, without the growing budgets of previous years.” “AI underpins how virtual care works at scale. Machine learning models are used to identify patients at risk of deterioration by combining NHS and proprietary datasets, while continuous data from clinical-grade wearables (e.g.oxygen saturation, blood pressure, ECG) is analysed to detect early warning signs. The lets clinical teams intervene sooner and safely manage far larger patient groups than would otherwise be possible.” Doccla and virtual care Doccla is a company providing remote patient monitoring and virtual wards to NHS trusts. The Doccla model is “designed both to support earlier discharge and to prevent avoidable admissions, particularly for those with long-term conditions.” There is already evidence for Doccla’s effectiveness, with the NHS seeing a 61% reduction in bed days, an 89% reduction in GP appointments, and a 39% drop in non-elective admissions. Not only has this AI-driven software improved efficiency, it is also reportedly saving the NHS approximately £450 a day compared with the cost of a hospital bed, the company says. Figures suggest that for every £1 spent on such technology, the NHS saves an estimated £3 compared with non-tech models. Mr Macdonnell said, “At Doccla, we use machine learning to identify patients at risk of deterioration before they reach crisis point. Continuous data from clinical-grade wearables like oxygen saturation, blood pressure and ECGs, are analysed with medical records to detect early warning signs.” The insights are allowing clinical teams to intervene sooner and manage larger caseloads compared with more traditional systems. AI may also be having a positive effect on clinician’s mental states, helping reduce administrative burden. For instance, large language models (LLMs) are being used to streamline clinical notes and present complex information to patients in a more accessible way. AI is not expected to replace clinicians, only make them more effective, so clinicians reading this can breathe a sigh of relief. Clinical trust in this technology remains low and this will only grow through transparency and further evidence of success. Predictive models must also deliver accurate and fair outcomes in diverse patient groups before being deployed at scale in real-world clinical settings. As the ***’s NHS works to move more care away from hospitals and into the community, with its “Fit for the Future: 10 Year Health Plan for England,” AI stands at the forefront of this transformation. The future of AI healthcare is set to allow patients to remain more independent and receive the care they need in familiar surroundings. (Image source: Pixabay under licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post AI helping ease the ***’s NHS burden appeared first on AI News. View the full article

Ahead of the AI & Big Data Expo at the San Jose McEnery Convention Center, May 18-19, we spoke to Jerome Gabryszewski, the company’s AI & Data Science Business Development Manager about AI, processing data for AI ingestion, and local versus cloud compute. The technology media is fond of quoting that data is ‘the new oil’, but the reality on the ground is that, despite having access to plenty of first-party information, actually leveraging it to the business’s advantage can prove problematic, especially at enterprise scale. Should you chose a cloud-hosted AI model, or local compute? How do you get your ‘data house’ in order, so the smart models can produce meaningful results? And as ever, we like to encourage our interviewees to help us predict the next chapter in the fast-moving story of business IT in this AI-dominated business landscape. Artificial Intelligence News: Moving from manual to automated data ingestion sounds great in theory, but it’s notoriously difficult. Where is HP seeing companies get stuck right now? One of the most consistent friction points we see is that organisations underestimate the organisational and architectural debt behind their data. Before automation can take hold, they have to reconcile fragmented data ownership across departments, inconsistent schemas in systems, and legacy infrastructure that was never designed for interoperability. The technical lift of automation is often smaller than the governance and integration work that has to precede it. Artificial Intelligence News: When AI models start updating themselves continuously, things can easily go sideways. How are you advising clients to handle risks like concept drift and data poisoning? Continuous learning is where AI goes from a project to a liability if it isn’t governed carefully. What we advise clients is to treat model updates the same way they treat code deployments. Nothing goes to production without a validation gate. For concept drift, that means MLOps pipelines with automated drift detection and human-in-the-loop triggers before retraining kicks in. For data poisoning, it’s a data provenance problem as much as a security problem. It’s critical to know exactly where your training data comes from and who can touch it. The clients who get this right aren’t necessarily the most technically sophisticated; It’s those who’ve embedded AI governance into their risk frameworks before they scaled. Artificial Intelligence News: I want to touch on HP’s hardware roots. What does a modern workstation or compute setup actually need to look like today to handle the sheer weight of an autonomous AI lifecycle? HP’s roots here actually matter. The Z series has been purpose-built for the most demanding professional compute for over 15 years so when we talk about what an autonomous AI lifecycle actually requires from hardware, we’re not guessing, we’ve been iterating on this problem longer than most! The answer isn’t a single machine, it’s a spectrum. At the individual developer level, you need local compute powerful enough to run real experiments without being cloud-dependent for every iteration. The ZBook Ultra and Z2 Mini handle the mobile and compact deskside tier professional-grade machines capable of running local LLMs and heavy workflows simultaneously. The ZGX Nano is where things get really interesting for AI-first teams. It’s an AI supercomputer that fits in the palm of your hand (15x15cm), but it’s powered by the NVIDIA GB10 Grace Blackwell Superchip with 128GB of unified memory and 1,000 TOPS of FP4 AI performance. A single unit handles models up to 200 billion parameters locally. And when a team needs to scale beyond that, you connect two units together via high-speed interconnect and you’re working with models up to 405 billion parameters… no cloud, no data centre, no ******. It comes pre-configured with the NVIDIA DGX software stack and the HP ZGX Toolkit, so teams go from setup to first workflow in minutes, not days. Moving up, the Z8 Fury gives power-user teams up to four NVIDIA RTX PRO 6000 Blackwell GPUs in a single system (384GB VRAM): That’s the full model development cycle running on-premises. And at the frontier, the ZGX Fury changes the conversation entirely. Powered by the NVIDIA GB300 Grace Blackwell Ultra Superchip with 748GB of coherent memory, it delivers trillion-parameter inference at the deskside, not the data centre. For teams running continuous fine-tuning and inference on sensitive data, it typically pays for itself in 8 to 12 months versus equivalent cloud compute. And for organisations that need to cluster and scale further, the entire Z portfolio is designed with rack-ready form factors that drop into managed IT environments without compromising security or data residency. Jerome Gabryszewski, AI & Data Science Business Development Manager, HP. The larger point is this; the autonomous AI lifecycle creates a governance and latency problem, not a compute problem. Teams can’t keep sending sensitive training data to the cloud every time a model needs to update. HP’s portfolio gives organisations a hardware path that scales with their workflow maturity, from the developer’s desk all the way to distributed on-premises compute. The hardware finally matches the ambition of what these AI systems actually need to do. Artificial Intelligence News: Gen AI compute costs are spiraling for a lot of enterprises. What is the practical fix for balancing that massive expense with modern cloud efficiency? The cost problem is structural, not cyclical. Enterprise GenAI spend surged to $37 billion in 2025, and 80% of companies still missed their cost forecasts by more than 25%. The core tension is that unit inference costs are actually falling, but total spend keeps rising because use is growing faster than cost drops. The cloud API model was designed for experimental, low-volume workloads. It was never built to be the economic engine for production AI at scale. The practical fix is a discipline problem before it’s an infrastructure problem: Draw a hard line between exploratory work and production workloads, and never use the same compute model for both. Early iterative work – prototyping, fine-tuning, model evaluation – should run on local hardware like the ZGX Nano or Z8 Fury, where you’re spending capital once instead of burning operational budget on experiments without a clear ROI path. The organisations getting this right are running a three-tier model: Cloud for burst training and frontier model access you’ve genuinely earned, on-premises HP Z infrastructure for predictable high-volume inference, and edge compute where latency is critical. Independent analysis shows on-premises can deliver up to an 18x cost advantage per million tokens over a five-year lifecycle. The framing we use with clients is simple: cloud is for scale you’ve earned, not scale you’re hoping for.” Artificial Intelligence News: Everyone wants their proprietary data to be ‘AI-ready.’ How do companies pull that off without exposing sensitive or siloed information? The mistake most companies make is treating ‘AI-ready data’ as a data engineering problem when it’s really a data sovereignty problem, and those require different solutions. Sending proprietary data to a cloud model for processing isn’t just an exposure risk, it’s a governance failure waiting to happen, especially in regulated industries where even the act of transmitting data externally can trigger compliance violations. The architecture that solves this is Retrieval-Augmented Generation (RAG) running on local infrastructure, which lets a model retrieve relevant context from your internal knowledge base at query time without ever training on it or exposing it externally. Your proprietary data stays on-premises, inside hardware you control. For example, a ZGX Nano or Z8 Fury running a locally hosted model can power a full RAG pipeline against sensitive internal documents with no data leaving the building and no token spend sent to a third party. The access control layer is where this gets operationally serious; a well-architected RAG system enforces role-based permissions at the retrieval level, so the AI surfaces only what a given employee is entitled to see, the same way your document management system does. The combination of local compute, local model, local retrieval, and governed access is what actually makes proprietary data AI-ready without exposure. The companies getting this right aren’t sending their crown jewels to the cloud to be processed; they’re bringing the intelligence to the data, not the other way around. Artificial Intelligence News: If we combine autonomous AI with these modern cloud platforms, what happens to the day-to-day role of an enterprise IT team over the next couple of years? I think Jensen Huang laid this concept out best. He said our job is not to wrangle a spreadsheet or type into a keyboard, that our work is generally more meaningful than that. And he’s drawn a sharp distinction between a job’s task and its purpose. In IT, for example, the task might be provisioning servers or triaging incidents, but the purpose is keeping the business resilient and moving forward. That distinction is exactly what’s playing out right now. Gartner projects 40% of enterprise applications will have embedded AI agents by end of 2026, up from less than 5% just a year ago, which means the routine execution layer of IT is being absorbed fast but the governance and architecture layer is expanding just as quickly. What’s already happening in leading organisations is a change from IT teams executing tasks to designing and governing the agents that execute on their behalf. The important gap is that only one in five companies has a mature governance model for that yet. This is where local-first infrastructure matters again. When your automation layer runs on hardware you control, you have full observability over agent behaviour that you simply don’t have when those workloads are abstracted into the cloud. The IT team of the next two years isn’t the team that keeps the lights on. It will be the teams that decide which agents get trusted with which decisions and makes sure the infrastructure underneath that judgement is something the business can actually stand behind. (Image source: Pixabay, licence.) Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo AI & Big Data Expo taking place in Amsterdam, California, and London. This comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post HP and the art of AI and data for the enterprise appeared first on AI News. View the full article

The US administration has added four more AI companies to its roster of favoured suppliers, with the Pentagon signing agreements with Microsoft, Reflection AI (which has yet to release a publicly-available model), Amazon, and Nvidia that mean their products can be used on classified operations. The companies join OpenAI, xAI, and Google as companies that the Department for Defense can deploy “for any lawful use.” The phrase “any lawful use” formed the centre of the recent disagreement between Anthropic AI and the US administration, with CEO Darius Amodei claiming that it would let the US government use Anthropic technology to subject the American civilian population to surveillance, and produce autonomous weapons, areas of Anthropic’s use that he wanted walled off. The Pentagon cancelled a $200 million contract with the company, a decision which Anthropic swiftly took to court, claiming millions in lost revenues from the government and others influenced by the government’s decision. The Trump administration termed the company a “supply chain risk”, the first time a US-based company had ever been given such a status. Ensuing statements from government sources described Anthropic as a “woke” company. The Pentagon’s statement on its new agreements reads, “The Department will continue to build an architecture that prevents AI vendor lock-in and ensures long-term flexibility for the Joint force.” The technologies will “give warfighters the tools they need to act with confidence and safeguard the nation against any threat.” The AIs will be used for ‘Impact Levels’ six (secret data) and seven (the most highly-classified materials) use-cases, helping create what the statement describes as an “AI-first fighting force”. The Pentagon’s current use of generative AI is largely confined to non-classified tasks carried out inside the various defence departments, such as working on document drafting and summary, and research. The new suppliers will help defence forces “streamline data synthesis” too, but also “elevate situational understanding, and augment warfighter decision-making in complex operational environments.” It’s not clear whether those descriptions include domestic deployments inside US borders. The expansion of the raft of AI suppliers to the US military and security forces means it will become more immune to apparent changes of heart by individual vendors affecting military and security operations. By broadening their technological base, the personal whims of individual company leaders become less relevant. Google and Amazon have in the past fired employees for protesting against their companies’ technology being used in weaponry and warfare. Anthropic’s Claude AI had been used on classified material as part of Palantir’s Maven toolset, a role which the most recent signees may replace. However, the company’s Mythos model is reportedly in use currently by the National Security Agency in the context of the platform’s purported cyber warfare and defence abilities. Worldwide, Anthropic’s Mythos is currently under assessment by 40 organisations, of which only 12 have been named, with the ***’s MI5 and the US NSA thought to be among the remaining 28. According to Axios, the US administration may be walking back on its most recent public stance on Anthropic. The website said it had a source in the White House who stated the administration was trying to find ways to “save face and bring ’em back in.” Anthropic’s Claude coding model is allegedly still in use by US government security organisations, and has been throughout recent events. According to the White House, the US government “continues to proactively engage across government and industry to protect our country and the American people, including by working with frontier AI labs.” (Image source: “BEST OF THE MARINE CORPS – May 2006 – Defense Visual Information Center” by expertinfantry is licensed under CC BY 2.0. Licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post US government increases AI suppliers and rethinks Anthropic’s role appeared first on AI News. View the full article

Google is testing Remy, a new AI personal agent for Gemini, according to Business Insider. The tool is designed to take actions for users in work and daily tasks. Remy is being tested in a staff-only version of the Gemini app. The report said it reviewed an internal document and spoke with two people familiar with the matter. The internal description presents Remy as a “24/7 personal agent”, intended to turn Gemini into an assistant that can act on a user’s behalf. Two people familiar with the project said Google employees are currently testing Remy. A Google spokesperson declined to comment. The report did not say when, or whether, Google plans to release Remy publicly. It also did not identify which Google services are included in the current employee test. Task-taking assistant Remy is part of Google’s broader work to expand Gemini beyond chat-based responses. Google already offers agent-related features, including Agent Mode, though access varies by subscription tier and region. The report described Remy as more advanced, and is designed to integrate in Google services and monitor things most relevant to users, handling complex tasks and learning user preferences. Gemini’s connected-app surface Google’s Gemini support documentation shows the current scope of Gemini’s connected services, which can connect with other services to complete user requests and provide more relevant responses. Connected Apps include Google Workspace services (Gmail, Calendar, Docs, Drive, Keep, and Tasks), and – according to. Google’s help documentation – GitHub, Spotify, YouTube Music, Google Photos, WhatsApp, Google Home, and Android utilities. Control questions Google’s Gemini Privacy Hub will give context, working with connected apps, including Google apps and third-party services. Users can review and delete Gemini Apps Activity, change auto-delete settings, and manage whether data is used to improve Google AI. It also lets users manage access to other apps and data, as well as information they have asked Gemini to save. Google’s existing Gemini documentation covers actions with different levels of user impact, including retrieving information from Workspace apps, creating calendar events, sending messages, opening apps, and controlling device or smart-home functions. Google Research says AI agents should have well-defined human controllers, carefully limited powers, observable actions, and the ability to plan. Google Cloud has also said agent activities should be transparent and auditable through logging and clear action characterisation. Its guidance emphasises limiting agent powers according to the intended purpose and user risk tolerance, using the least-privilege principle. Remy’s reported preference-learning function also puts memory controls in focus. Google’s Privacy Hub says users can manage information they have asked Gemini to save and covers controls for personalisation based on past chats and Personal Intelligence. The report did not provide technical details on Remy’s architecture, the model version behind it, or the level of autonomy being tested. It also did not say whether Remy can act independently without user confirmation. Those unanswered points mean it’s unclear how Remy handles approvals and logs completed-action. The internal document describes Remy as a dog-fooding project, a term commonly used in technology companies when employees test products before any broader release. The report compared Remy’s concept with OpenClaw, an AI agent that drew attention earlier this year for its ability to autonomously reply to messages, conduct research on behalf of users, and take autonomous actions. OpenAI CEO Sam Altman said in February that OpenAI was hiring OpenClaw’s creator, according to the report. Google DeepMind CEO Demis Hassabis has previously discussed the goal of building a digital assistant, but Google has not confirmed whether Remy will become a public Gemini feature. (Photo by Kai Wenzel) See also: Google made agentic AI governance a product. Enterprises still have to catch up. Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Google tests Remy AI agent for Gemini as focus turns to user control appeared first on AI News. View the full article

Governance around Physical AI is becoming harder as autonomous AI systems move into robots, sensors, and industrial equipment. The issue is not only whether AI agents can complete tasks. It is how their actions are tested, monitored, and stopped when they interact with real-world systems. Industrial robotics already provides a large base for that discussion. The International Federation of Robotics said 542,000 industrial robots were installed worldwide in 2024, more than double the annual level recorded a decade earlier. It expects installations to reach 575,000 units in 2025 and pass 700,000 units by 2028. Market researchers are also applying the Physical AI label to a wider group of systems, including robotics, edge computing, and autonomous machines. Grand View Research estimated the global Physical AI market at US$81.64 billion in 2025 and projected it to reach US$960.38 billion by 2033, though the category depends on how vendors define intelligence in physical systems. From model output to physical action The governance challenge is different from software-only automation because physical systems can operate around workplaces, infrastructure, and human users. They can also be connected to equipment that requires clear safety limits. A model output can become a robot movement or a machine instruction. It can also become a decision based on sensor data. That makes safety limits and escalation paths part of system design. Google DeepMind’s robotics work is one recent example of how AI models are being adapted for this environment. The company introduced Gemini Robotics and Gemini Robotics-ER in March 2025, describing them as models built on Gemini 2.0 for robotics and embodied AI. Gemini Robotics is a vision-language-action model designed to control robots directly, while Gemini Robotics-ER focuses on embodied reasoning, including spatial understanding and task planning. A robot using this type of model may need to identify an object, understand an instruction, and plan a sequence of movements. It also needs to assess whether the task has been completed correctly. That creates a control problem that includes both model behaviour and the mechanical limits of the system. Google DeepMind said useful robots need generality, interactivity, and dexterity. Generality covers unfamiliar objects and environments. Interactivity relates to human input and changing conditions. Dexterity refers to physical tasks that require precise movement. In its launch materials, Google DeepMind said Gemini Robotics could follow natural-language instructions and perform multi-step manipulation tasks. Examples included folding paper, packing items into a bag, and handling objects not seen during training. The technical requirements for Physical AI are broader than language understanding. Systems need visual perception and spatial reasoning. They also need task planning and success detection. In robotics, success detection matters because the system must decide whether a task has been completed, whether it should retry, or whether it should stop. Google DeepMind’s Gemini Robotics-ER 1.6, introduced in April 2026, shows how those functions are being packaged in newer models. The company describes the model as supporting spatial logic, task planning, and success detection, with the ability to reason through intermediate steps and decide whether to move forward or try again. Google’s developer documentation says Gemini Robotics-ER 1.6 is available in preview through the Gemini API. The documentation describes it as a vision-language model that brings Gemini’s agentic capabilities to robotics. Those capabilities include visual interpretation, spatial reasoning, and planning from natural-language commands. Google AI Studio provides a developer environment for working with Gemini models, while the Gemini API provides a route for integrating those models into applications. In the context of embodied AI, that places testing and prompting closer to the developers building agentic applications. Safety controls move into system design Governance becomes more complex when these systems can call tools, generate code, or trigger actions. Controls need to define what data the system can access, what tools it can use, which actions require human approval, and how activity is logged for review. McKinsey’s 2026 AI trust research points to the same issue in enterprise AI more broadly. It found that only about one-third of organisations reported maturity levels of three or higher in strategy, governance, and agentic AI governance, even as AI systems take on more autonomous functions. In robotics, safety also includes the physical behaviour of the machine. Google DeepMind has described robot safety as a layered problem, covering lower-level controls such as collision avoidance, force limits, and stability, as well as higher-level reasoning about whether a requested action is safe in context. The company also introduced ASIMOV, a dataset for evaluating semantic safety in robotics and embodied AI. Google DeepMind said the dataset was designed to test whether systems can understand safety-related instructions and avoid unsafe behaviour in physical settings. The same controls used for software agents become harder to manage when systems are connected to robots, sensors, or industrial equipment. These include access rights, audit trails, and refusal behaviour. They also include escalation paths and testing. Governance frameworks such as the NIST AI Risk Management Framework and ISO/IEC 42001 provide structures for managing AI risks and responsibilities across the system lifecycle. In Physical AI, those controls need to account for model behaviour, connected machines, and the operating environment. Google DeepMind has also worked with robotics companies as part of its embodied AI development. In March 2025, the company said it was partnering with Apptronik on humanoid robots using Gemini 2.0, and listed Agile Robots, Agility Robotics, Boston Dynamics, and Enchanted Tools among trusted testers for Gemini Robotics-ER. The 2026 update also referenced work with Boston Dynamics involving robotics tasks such as instrument reading. That type of use case depends on visual understanding, task planning, and reliable assessment of physical conditions. Physical AI applies to industrial inspection, manufacturing, and logistics. It also applies to facilities and warehouses. These settings require systems to interpret real-world conditions and act within defined limits. The governance question is how those limits are set before autonomous systems are allowed to make or execute decisions. Google DeepMind and Google AI Studio are listed as hackathon technology partners for AI & Big Data Expo North America 2026, taking place on May 18–19 at the San Jose McEnery Convention Center. (Photo by Mitchell Luo) See also: AI agent governance takes focus as regulators flag control gaps Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Physical AI raises governance questions for autonomous systems appeared first on AI News. View the full article

Two weeks ago at Google Cloud Next ’26 in Las Vegas, Google did something the enterprise AI industry has been dancing around for the better part of two years: it made agentic AI governance a native product feature, not an afterthought. The centrepiece announcement was the Gemini Enterprise Agent Platform, pitched as the successor to Vertex AI and described by Google as a comprehensive platform to build, scale, govern, and optimise agents. What made it notable wasn’t the model access or the TPU upgrades, significant as those are. It was the architecture underneath: every agent built on the platform gets a unique cryptographic identity for traceability and auditing, while Agent Gateway handles oversight of interactions between agents and enterprise data. Governance, in other words, ships with the product. That design choice is a direct response to a problem that has quietly been undermining enterprise AI deployments across the board. The governance gap that no one wants to talk about A survey of 1,879 IT leaders by OutSystems, released in April, puts the numbers plainly: 97% of organisations are already exploring agentic AI strategies, and 49% describe their own capabilities as advanced or expert. Yet only 36% have a centralised approach to agentic AI governance, and just 12% use a centralised platform to maintain control over AI sprawl. That is an 85-point gap between confidence and actual control, and it is not improving fast enough. Gartner’s 2026 Hype Cycle for Agentic AI frames the same tension differently. Only 17% of organisations have actually deployed AI agents to date, yet more than 60% expect to do so within two years, the most aggressive adoption curve Gartner has recorded for any emerging technology in the survey’s history. The hype cycle places agentic AI squarely at the Peak of Inflated Expectations, with governance, security, and cost-management capabilities still maturing well behind deployment intent. The production reality is considerably more sobering. Multiple independent analyses put the share of agentic AI pilots that have reached genuine production scale at somewhere between 11% and 14%. The rest, the other 86% to 89%, have stalled, been quietly shelved, or never moved beyond proof-of-concept. Governance breakdowns and integration complexity are consistently cited as the primary causes, ahead of any technical shortcomings in the models themselves. What Google is actually betting on At Cloud Next ’26, the message from Google was less about model capability and more about who owns the control plane. Bain & Company’s post-event analysis noted that Google is repositioning from model access toward a full agentic enterprise platform, one where context, identity, and security sit at the centre of the architecture, not at the edges. The strategic logic is coherent. All three major cloud providers only announced agent registries in April 2026, which signals just how early-stage the governance tooling still is across the industry. Google’s move is the most comprehensive response so far, but it also carries a specific implication for enterprises evaluating the platform: deeper integration with Google’s stack is part of the deal. That tension–between the genuine governance capabilities on offer and the platform commitment required to access them–is what enterprise architects are now working through. Agentic systems multiply identities and permissions at a pace that traditional human-centric identity and access management models were never built to handle. Once agents start acting across systems, the governance question shifts from which model is approved to what actions a given agent can take, through which identity, against which tools, and with what audit trail. Google’s cryptographic agent identity and gateway architecture is a direct answer to that question. Whether enterprises are ready to hand Google that level of operational centrality is a different conversation. Agent washing makes this harder There is a compounding problem that the governance debate tends to sidestep: a large share of what is currently being marketed as agentic AI is not agentic AI. Deloitte’s research on enterprise AI trends notes that many so-called agentic initiatives are actually automation use cases in disguise: legacy workflow tools with conversational interfaces, operating on predefined rules rather than reasoning toward goals. The distinction matters because governance frameworks designed for genuinely autonomous agents will not map cleanly onto scripted automation, and vice versa. Enterprises that conflate the two end up with governance structures that are either too restrictive for real agents or too permissive for brittle automation masquerading as intelligence. Gartner estimates that more than 40% of agentic AI projects could be cancelled by 2027, with unclear value and weak governance cited as the leading reasons. That figure should concentrate minds. The enterprises investing now in governance architecture–audit trails, escalation paths, bounded autonomy, agent-level identity–are building the foundation that will determine whether their agentic deployments survive contact with production. Google’s Cloud Next platform launch is, at minimum, a forcing function. The tooling for governed agentic systems now exists at scale from a major provider. What remains is the harder organisational work–deciding what agents are actually authorised to do, who is accountable when they get it wrong, and whether the platform holding all of that together is one you are prepared to build on. See also: SAP: How enterprise AI governance secures profit margins Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Google made agentic AI governance a product. Enterprises still have to catch up. appeared first on AI News. View the full article

According to SAP, enterprise AI governance secures profit margins by replacing statistical guesses with deterministic control. Ask a consumer-grade model to count the words in a document, and it will often miss the mark by ten percent. Manos Raptopoulos, Global President of Customer Success Europe, APAC, Middle East & Africa at SAP, observes that the operational gap between near-perfect and perfect is absolute. “The distance between 90% and 100% accuracy is not incremental. In our world, it is existential,” notes Raptopoulos. As organisations push large language models into production environments, Raptopoulos emphasises that the evaluation criteria have formally transitioned toward precision, governance, scalability, and tangible business impact. The pressing challenge facing corporate boards centres on the evolution from passive tools to active digital actors, a transition Raptopoulos identifies as the primary governance moment and will be among the topics that SAP will be focusing on at this year’s AI & Big Data Expo North America. Agentic AI systems now possess the capability to plan, reason, orchestrate with other agents, and execute workflows autonomously. Because these systems interact directly with sensitive data and influence decisions at scale, Raptopoulos argues that failing to govern them exactly as one governs a human workforce exposes the organisation to severe operational risk. He warns that agent sprawl will mirror the shadow IT crises of the past decade, though the stakes are categorically higher. Establishing agent lifecycle management, defining autonomy boundaries, enforcing policy, and instituting continuous performance monitoring are mandatory requirements, according to his framework. Integrating modern vector databases (which map the semantic relationships of enterprise language) with legacy relational architectures demands immense engineering capital. Teams must actively restrict the agent’s inference loop to prevent hallucinations from corrupting financial or supply chain execution paths. Setting these strict parameters drives up computational latency and hyperscaler compute costs, altering initial P&L projections. When an autonomous model requires constant, high-frequency database querying to maintain deterministic outputs, the associated token costs multiply quickly. Governance becomes a hard engineering constraint rather than a compliance checklist. Raptopoulos argues that corporate boards must resolve three baseline issues before deploying agentic models: identifying who holds accountability for an agent’s error, establishing audit trails for machine decisions, and defining the exact thresholds for human escalation. Geopolitical fragmentation makes answering these questions harder. Sovereign cloud infrastructures, AI models, and data localisation mandates are regulatory realities in major markets spanning New York, Frankfurt, Riyadh, and Singapore. Enterprises must embed deterministic control directly into probabilistic intelligence. Raptopoulos views this requirement as a C-suite mandate rather than an IT project. Structuring relational intelligence for commercial operations AI systems remain entirely dependent on the quality of the data and processes they operate upon, representing what Raptopoulos calls the data foundation moment. Fragmented master data, siloed business systems, and over-customised ERP environments introduce dangerous unpredictability at the worst possible moments. Raptopoulos explains that if an autonomous agent relies on fragmented foundations to provide a recommendation affecting cash flow, customer relations, or compliance positions, the resulting operational damage scales instantly. Extracting tangible enterprise value requires advancing beyond generic large language models trained on internet-scale text. True enterprise intelligence – as outlined by Raptopoulos – must be grounded in proprietary corporate data, including orders, invoices, supply chain records, and financial postings embedded directly into business processes. He argues that relational foundation models optimised specifically for structured business data will continually outperform generic models in forecasting, anomaly detection, and operational optimisation. The sheer operational friction of making an over-customised ERP environment intelligible to a foundation model halts many deployments. Data engineering teams spend excessive cycles sanitising fragmented master data simply to create a baseline for the AI to ingest. When a relational model needs to accurately interpret complex, proprietary supply chain records alongside raw invoice data, the underlying data pipelines must operate with zero latency. If the data ingest fails, the model’s predictive capabilities degrade instantly, rendering the agent functionally dangerous to the business. Integrating legacy architecture with modern relational AI requires overhauling deeply entrenched data pipelines. Engineering teams face indexing decades of poorly classified planning data so that embedding models can generate accurate vector representations. Following Raptopoulos’s logic, boards must evaluate whether their current data estate is genuinely prepared, rather than simply layering probabilistic intelligence over disjointed foundations. Designing intent-based interfaces Enterprise application interaction is transitioning from static interfaces to generative user experiences, a development Raptopoulos flags as the employee interaction moment. Instead of manually navigating complex software ecosystems, employees will express their intent to the system. Raptopoulos offers the example of a user instructing the software to prepare a briefing for their highest-revenue customer visit that week. The AI agents then orchestrate the necessary workflows, assemble the surrounding context, and surface recommended actions. However, Raptopoulos stresses that adoption among the workforce remains conditional upon trust. Employees will only embrace these digital teammates when they feel confident that the system’s outputs respect established governance boundaries, reflect authentic business rules, and deliver demonstrable productivity gains. Engineering these systems demands role-specific AI personas tailored for positions such as the CFO, the CHRO, or the head of supply chain. Raptopoulos observes that these personas must be built upon trusted data and embedded within familiar corporate workflows to successfully close the adoption gap. Achieving this level of integration is a design decision carrying heavy consequences. Organisations willing to invest capital into AI-native architecture accelerate their return on investment, while enterprises attempting to bolt probabilistic models onto legacy interfaces struggle heavily with trust, usability, and scale. Technology leaders trying to force modern AI orchestration onto monolithic software applications often encounter severe integration delays. The routing of probabilistic API calls through outdated enterprise middleware causes user interfaces to lag, destroying the intent-based workflow. Designing role-specific personas requires more than prompt engineering; it demands mapping complex access controls, permissions, and business logic into the model’s active memory. Engineering competitive defense The financial return on AI surfaces fastest during customer interactions. Raptopoulos notes that training models on proprietary records, internal rules, and historical logs creates a layer of customer-specific intelligence that rivals cannot easily copy. This setup performs best in exception-heavy workflows like dispute resolution, claims, returns, and service routing. Deploying autonomous agents capable of classifying cases, surfacing relevant documentation, and recommending policy-aligned resolutions converts these high-cost processes into distinct competitive differentiation. These models adapt based on the results of each interaction. Raptopoulos points out that corporate buyers prioritise reliable, relevant, and responsive service rather than technological gimmicks. Companies that deploy AI to handle heavy workloads – while maintaining strict oversight of the final outputs – construct barriers to entry that generic tools fail to penetrate Deploying corporate intelligence requires the C-suite to orchestrate three distinct layers in parallel, which Raptopoulos defines as the strategy moment. The initial layer involves embedded functionality, where persona-driven productivity gains are integrated directly into core applications for fast returns. The second layer demands agentic orchestration, facilitating multi-agent coordination across cross-system workflows. The final layer focuses on industry-specific intelligence, featuring deeply specialised applications co-developed to address the highest-value challenges specific to a particular sector. A trap awaits leaders who fall victim to false sequencing. Concentrating solely on embedded tools leaves massive financial value uncaptured, while jumping aggressively toward deep industry applications without first achieving proper governance and data maturity multiplies corporate risk. Raptopoulos advises that scaling these models requires matching corporate ambition to actual technical readiness. Leadership teams need to fund clean core architectures, update data pipelines, and enforce cross-functional ownership to move past the pilot phase. The most profitable deployments treat AI as a central operating layer that requires the same governance as human staff. The financial gap between 90 percent accuracy and full certainty dictates where true enterprise value lives. Governance decisions made in the coming months will dictate whether specific AI deployments become a powerful source of durable advantage, or an expensive lesson. See also: AI agent governance takes focus as regulators flag control gaps Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post SAP: How enterprise AI governance secures profit margins appeared first on AI News. View the full article

As of 1st June 2026, GitHub Copilot will charge its users on the basis of the tokens they use, rather than a flat rate subscription model. The model that’s seeing the shutters closed on it is, or rather was, simple to understand and use. Users were given a set number of ‘Premium Requests’ according to their subscription tier. A complex coding task that may have taken many hours to complete used one premium request. Posing a relatively trivial question also counted as a single premium request. However, the change which is soon to affect GitHub Copilot users aligns the pricing models with those of API charges to large language models, more common among business plans. On the new GitHub Copilot pricing scheme, most requests will be measured according to the tokens used by, input to, and output from the LLM at the heart of Copilot. The definition and cost of tokens A token is often described as representing around three-quarters of a word. Thus, giving an LLM a text of 10,000 words to examine would equate to 12,000-13,000 tokens of content. In developer terms, if a body of code which Copilot were to examine (for refactoring or bug-hunting for example), comprised of 10,000 ‘words’ (expressions, statements, variable names, functions, and so on), then that using it in one query, once, would count as 12,000-13,000 tokens out of their allottment for the month. Prompt text, as inputs, will also count, as will the outputs from Copilot. The pricing tiers coming into effect next month remain pegged at their current levels, but instead of being allotted a number of queries per month, users are given ‘AI Credits’ to the same value. A base-tier Copilot Pro subscriber ($10pcm) will receive 1,000 credits, with GitHub saying that at present one AI Credit is worth one US cent. The number of tokens each credit buys will depend on the model used, the input/output mix, the size of the cache (data held in the LLM’s memory for context), and feature requested. Thus, if a developer uses mostly simple queries, they are likely not to have to buy extra tokens in the form of credits each month. Conversely, multi-agent queries about a complex, lengthy code base will empty the AI Credit account more quickly. Queries to the most-advanced frontier models will cost more than to the less-powerful. GitHub’s pricing changes do include some compensatory benefits for users: Code completions (similar to a phone’s auto-complete function) and Next Edit suggestions will remain free. The industry changes to per-token pricing The changes to GitHub’s pricing model are in line with similar changes from other companies. Anthropic and OpenAI have now moved their enterprise customers to token-based billing. Unlike those two, however, Microsoft – owner of GitHub – is a profitable business overall, and has to date been able to subsidise the use of GitHub Copilot with revenues from other parts of the business, such as its software and cloud divisions. Up until the change on 1st June, users will have been able to ‘spend’ between three and eight times the number of tokens their monthly subscription costs have covered, and incurred no penalty. Microsoft’s move is a change that affects those it was hoping to attract to Copilot’s features, immediately forcing new and existing users to become aware of their token spend per query – a figure that has been abstracted away by per-month subscriptions to date. The new billing model may make more economic sense from Microsoft’s point of view, but it discourages the exploration and testing that new users will want to do. For businesses that deploy AI coding agents in their development teams, the cost implications of the industry-wide shift in pricing policies are significant. In the case of Uber, for instance per The Information [paywall], its CTO has said it had spent the year’s AI budget for 2026 already this year, pointing out that 11% of updates to Uber’s code are now written by AI. Uber primarily uses Anthropic’s Claude coding agents. Outside the IT department, companies deploying AI automation should be aware that complex tasks, which may involve running agentic LLMs unsupervised for long periods, could soon be charged on a similar per-token basis. Thus, the delivered efficiency gains from AI in the workforce will have to be measured against any rise in AI vendors’ bills. (Image source: Pixabay under licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post Per-token AI charges come to GitHub Copilot appeared first on AI News. View the full article

LG is currently engaged in exploratory discussions with NVIDIA concerning physical AI, data centres, and mobility. Following a meeting in Seoul between LG CEO Ryu Jae-cheol and Madison Huang, Senior Director of Product Marketing for Omniverse and Robotics at NVIDIA, the core operational dependencies required to run complex automated systems are becoming apparent. While the companies have not formalised investment amounts or timelines, their intersecting hardware and processing priorities highlight the massive capital expenditure required to bring autonomous systems out of simulation. The densification of compute clusters required for complex machine learning models creates an unavoidable physics problem. NVIDIA’s data centre business generates record revenues, but operating these high-density server racks pushes conventional cooling infrastructure past safe operating limits. At CES 2026, LG positioned its commercial divisions to supply high-efficiency HVAC and thermal management solutions engineered for AI data centres. As power density explodes in relevance, traditional air cooling is simply inadequate. When server farm temperatures exceed safe thresholds, compute nodes throttle performance, destroying the return on investment for high-end silicon. Integrating LG’s thermal hardware directly into NVIDIA’s infrastructure ecosystem addresses this margin drain. It allows facility operators to pack more processing power into smaller square footage without burning out the underlying hardware. For LG, this positions them as an infrastructure supplier inside a lucrative technology ecosystem, generating recurring enterprise revenue by complementing the compute layer rather than competing against it. Underscoring this broader push into connected enterprise systems, LG subsidiary LG CNS is a sponsor of this year’s IoT Tech Expo North America, signaling the company’s aggressive expansion across smart infrastructure. Hardware actuation and edge inference friction Beyond server infrastructure, the discussions attempt to solve the computational latency inherent in autonomous consumer hardware. LG’s future growth thesis relies heavily on automating household manual and cognitive workloads. LG recently unveiled CLOiD, a home robot featuring two arms with seven degrees of freedom and five individually-actuated fingers per hand. This hardware runs on LG’s ‘Affectionate Intelligence’ platform, built for contextual awareness and continuous environmental learning. Translating a computational command into physical movement requires a flawless zero-latency inference pipeline. When an articulated robot reaches for a glass, the system must process real-time visual data, query local vector databases to identify the object’s properties, and calculate the exact required grip force. Any miscalculation within this inference pipeline risks physical damage to the user’s home. LG currently lacks the digital twin infrastructure, pre-trained manipulation models, and simulation environments necessary to compress this deployment pipeline securely. NVIDIA provides this architecture through its Omniverse and Isaac robotics stack, which are optimised for real-time physical AI inference. By adopting NVIDIA’s edge-compute capabilities, LG can process complex spatial variables locally, heavily reducing the cloud compute costs associated with continuous spatial mapping and video ingestion. This proven pipeline compresses the time required to move from prototype to full commercial production. Mass market ingestion and simulation environments NVIDIA is concurrently validating its robotics stack, having wrapped a two-week Siemens factory trial in January 2026 that was just announced at Hannover Messe in April. During this trial, a Humanoid HMND 01 Alpha executed live logistics operations over an eight-hour *******. Yet, factory floors in Erlangen are highly structured and regulated. Consumer living rooms contain extreme variability, changing lighting, and unpredictable human interference. Accessing LG’s ThinQ ecosystem and its mass-market distribution provides NVIDIA with a data-rich training environment. Bringing robots into homes requires training models on actual domestic variability rather than sterile simulations. Moving beyond industrial settings into consumer electronics gives NVIDIA’s Omniverse platform the potential to become the universal development infrastructure for real-world autonomy, mirroring how its GPU architecture captured cloud processing. The final alignment point covers automotive integration. LG’s automotive components division represents one of its fastest-growing segments, manufacturing in-vehicle infotainment, EV components, and in-cabin generative platforms that include gaze-tracking and adaptive displays. Simultaneously, NVIDIA’s DRIVE platform commands massive deployment share in autonomous and semi-autonomous vehicle computing. Automotive manufacturers frequently struggle when attempting to bridge legacy infotainment systems with advanced autonomous compute nodes. Because LG and NVIDIA already operate in adjacent layers of the same vehicle, a formal collaboration would unite LG’s interior experience layer with NVIDIA’s underlying compute platform. This unification allows fleet operators to standardise their reference architectures, reducing the engineering hours wasted on custom API integrations and securing a unified pathway for over-the-air machine learning updates. These exploratory talks between LG and NVIDIA define the precise hardware and processing requirements necessary to execute physical AI reliably. See also: Kakao Mobility details Level 4 autonomous driving roadmap for physical AI Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post What LG and NVIDIA’s talks reveal about the future of physical AI appeared first on AI News. View the full article

APIs and MCPs are often mentioned in the same breath as ways that systems can exchange information, but they are designed differently and have different purposes. This article hopes to explain the differences and how software developers and users should approach interaction with each. An API is mainly found in software applications, while an MCP (Model Context Protocol), is used by large language models. APIs let one application talk to another, and an MCP lets an AI model use data and tools in structured ways. The difference comes about because LLMs, responding to user requests, need to choose which tools and information it thinks it needs to achieve an outcome. APIs: Simple definition An API sends a request in an agreed format to another software instance, and receives a response in the agreed format, with the details of each exchange’s protocols (or methods of behaviour) hard-coded. Developers write code to call out to an API and create code to parse, or handle, the response. This makes APIs precise and reliable – although the interchange can falter if either party changes the code governing the API’s behaviour. APIs are still important to systems using LLMs, and many AI-based systems rely on APIs to function. A model may request data, and get responses via an API. MCPs: Simple definition MCPs are used when LLMs need access to data in situations like needing to query business data repositories, read the contents of particular files, or trigger an action. MCPs give models a structured way to access multiple data sources via one interface. An MCP server exposes data in a standard format according to rules set up in advance. These rules determine what is available and to whom or what. MCP servers expose three kinds of ability: Tools are actions the model may instigate, like creating a file or searching a database. Resources are information the model may read as context. Prompts are reusable templates that help users perform common tasks, without having to write a detailed prompt every time they perform the same action. The important difference is that MCPs are designed for a model to be the direct consumer of data. The model suggests which tools or resources it requires according to what it thinks may be relevant to the user’s request. Why MCPs are not an API wrappers In some systems, APIs remain in use, but have an MCP placed between them and the user. An MCP server might call an API ‘behind the scenes’. However, an API could return more information by default than a model needs to achieve a task. But as every byte of data will need to be processed by the LLM, this can burn through many more tokens than are necessary. Too much information increases costs and can make the model’s answer less accurate. For example, an API might return 50 database fields about a customer, but the LLM requires a single account status entry. Sending all 50 fields gives the model more to process, which doesn’t necessarily provide useful context. The LLM has no idea of the relevance of the data until it has used processing cycles to determine the fact. Additionally, it may base its responses on extraneous data it’s been given, and produce inaccurate answers. In an ideal scenario, MCP tools are designed around the tasks a model needs to complete. If the user asks how many customers are subscribed to a particular service, or have bought a specific item, for example, the MCP tool will return the relevant numbers, rather than complete customer interaction records. When each are used Use an API when one application needs to communicate with another application when there is full knowledge between both parties as to what information is required. A website, mobile app, internal system, payment platform, or reporting tool will often use APIs. If the end-consumer of data is an AI model that needs access to undefined information or actions, an MCP should be used. An AI assistant that answers staff questions (with variable input, therefore) or is tasked to review internal documents may use MCPs. In many organisations, both exist. A customer app that can present specific information (an account balance, for instance) may call APIs. An AI assistant in the same app may use an MCP server because the nature of the queries it will create on behalf of the user will vary. Both may reach the same underlying data, but do so through different interfaces according to the type of system asking. Security and gateways A gateway is a device (usually instantiated in software) that fronts both types of service. It handles authentication, rate limits, logging, monitoring, and access control. If MCP use grows, organisations need to know which AI tools are requesting data from which systems, what data they are allowed access to, and what actions they can perform on that data. A gateway can create a place to manage these types of controls. However, as they operate at the network layer (arbitrating and recording data movement), they do not solve problems that emanate from the software layer (including LLMs, deterministic code, or user activity). In cybersecurity terms, they can be thought of as a firewall: useful in certain contexts, but like firewalls, they can be circumvented, represent a single point of failure, and might give a false sense of security. MCP and API gateways are arguably perimeter defences, that will not reliably prevent data-related incidents. These are still possible when caused by software, either deterministic, ‘traditional’ code or an LLM. (Image source: Pixabay under licence.) Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and co-located with other leading technology events. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post A guide to APIs, MCPs, and MCP Gateways appeared first on AI News. View the full article

Australia’s financial regulator has warned financial firms that AI agent governance and assurance practices are poorly governed. The warning comes as banks and superannuation trustees expand AI in internal and customer-facing operations. The *********** Prudential Regulation Authority said it conducted a targeted review of selected large regulated entities in late 2025 to assess AI adoption and related prudential risks. It found that AI was being used in all entities reviewed, but maturity varied in risk management and operational resilience. APRA said boards showed strong interest in AI for productivity and customer experience. However, it found that many were still building management of AI risks. The regulator also raised concerns about reliance on vendor presentations and summaries. It said boards were not always giving enough scrutiny to risks like unpredictable model behaviour and the effect of AI failures on critical operations. APRA said boards should develop a better understanding of AI in order to set strategy and oversight coherently. It said AI strategy should align with an institution’s risk appetite and include monitoring and defined procedures that should be taken in the event of errors. APRA noted regulated entities were trialling or introducing AI in software engineering, claims triage, and loan application processing. Other use cases cited included fraud and scam disruption and customer interaction. Some entities were treating AI risk in the same terms as that of other technologies, but that approach doesn’t account for models’ behaviour and bias. It identified gaps in model behaviour monitoring, change management, and decommissioning, and stated a need for inventories of AI tools and named-person ownership of AI instances. It also pointed out the requirement for human involvement in high-risk decisions. Cybersecurity was another area of concern. APRA said AI adoption was changing the threat environment by adding additional attack pathways such as prompt injection and insecure integrations. Identity and access management practices had not adjusted in some instances to non-human elements such as AI agents. The volume of AI-assisted software development was placing pressure on change and release controls. APRA said entities should apply controls on agentic and autonomous workflows which included privileged access management, configuration, and patching. It also called for security testing of AI-generated code. Some institutions had become dependent on a single provider for many of their AI instances, ARPA noted, and only a few had been able to show an exit plan or substitution strategy for AI suppliers. APRA said AI can be present in upstream dependencies, which entities may not be aware of. Identity and access The focus on identity and permission controls is also reflected in new standards work by the FIDO Alliance. The group has formed an Agentic Authentication Technical Working Group and is developing specifications for agent-initiated commerce. FIDO said some existing authentication and authorisation models were designed for human interaction, not delegated actions performed by software. It said service providers need ways to verify who or what authorises actions and under what conditions. Vendors have presented their solutions to FIDO for review, including Google’s Agent Payments Protocol and Mastercard’s Verifiable Intent framework. The Centre for Internet Security, a non-profit funded largely by the Department for Homeland Security, has published AI security companion guides that map CIS Controls v8.1 to large language models, AI agents, and Model Context Protocol environments. Its LLM guide covers prompt and sensitive-data issues, and an MCP guide focuses on secure access by software tools, non-human identities, and network interactions. (Photo by julien Tromeur) See also: Google warns malicious web pages are poisoning AI agents Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information. AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here. The post AI agent governance takes focus as regulators flag control gaps appeared first on AI News. View the full article