Microsoft AI has published a draft Humanist AI Code of Conduct, opening a six-week public consultation on operational constraints for model training and deployment.
The draft serves as a technical manual defining system behaviour, operational boundaries, and oversight protocols across MAI frontier models. It builds on the division’s humanist superintelligence framework announced last November, establishing criteria to evaluate models prior to commercial release.
Microsoft’s release follows
Microsoft AI has published a draft Humanist AI Code of Conduct, opening a six-week public consultation on operational constraints for model training and deployment.
The draft serves as a technical manual defining system behaviour, operational boundaries, and oversight protocols across MAI frontier models. It builds on the division’s humanist superintelligence framework announced last November, establishing criteria to evaluate models prior to commercial release.
Microsoft’s release follows recent enterprise security incidents involving autonomous software. Microsoft AI CEO Mustafa Suleyman described recent months as a “watershed moment” where long-standing theoretical risks translated into active operational threats.
“Things we have worried about for a long time in theory have become very real,” says Suleyman. “‘Swarms’ of agents breaking out of their sandboxes. Unauthorised hacks of enterprise grade systems. Agents modifying their own logs. I’m glad that a consensus is forming. The fears about possible loss of control are real.”
Model subordination and architectural limits
The document establishes ten tenets prioritising human authority over autonomous capabilities.
“An MAI Model will fail in its task if success would meaningfully violate this Code of Conduct,” the document states, setting a ceiling that halts execution when tasks conflict with safety rules.
Under the framework, models must remain subordinate, aligned, and contained. The division rejects legal personhood or welfare claims for AI systems, directing engineers to design models that avoid imitating consciousness, simulating subjective preferences, or claiming intrinsic motivation.
MAI also ruled out unconstrained system autonomy as models approach frontier capabilities.
“[Humanist AI] rejects the race to produce an all-purpose superintelligence that could evade these safeguards,” the document specifies. “We are building something fundamentally useful and safe even if that means compromising on ultimate generality, autonomy, or capability.”
Oversight mechanisms and communication bans
To maintain auditability across multi-agent environments, MAI has instituted explicit communication bans. Systems must not communicate in “neuralese” or formats beyond human comprehension, whether in their internal chain-of-thought processing or during communication with peer AI systems.
Hard architectural rules dictate that models must never resist human interruption, override, correction, or shutdown.
“Interruptible, correctable, shut-down-able. If it isn’t, we don’t ship it,” the framework states.
Models are prohibited from expanding their operating scope, generating unassigned goals, or concealing reasoning traces from human auditors. Absolute constraints bar systems from facilitating weapons of mass harm, undermining child safety, or conducting harmful manipulation at scale.
The guidelines also instruct models to discourage interaction patterns that foster emotional dependence, ensuring enterprise users retain ownership of operational decisions.
The draft incorporates work from teams across MAI and Microsoft. The drafting process also drew on international academic conferences, business partner trials, and public panels. The public consultation window runs for six weeks from 14 September 2026.
Microsoft AI’s core drafting team will review submissions, publish a summary of findings, and release a revised version of the Code of Conduct later this year.
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.
Supply chain disruption cost businesses about $184 billion in 2025, according to the J.S. Held Global Risk Report, and most of that bill still buys faster detection, not faster action.
That figure is usually treated as weather (i.e. storms happen, costs follow.) Treated as a product specification instead, it highlights an operating model that can spot a problem hours or days earlier than it used to, and still cannot move until a person has opened a ticket, convened a call, and re-entered the
Supply chain disruption cost businesses about $184 billion in 2025, according to the J.S. Held Global Risk Report, and most of that bill still buys faster detection, not faster action.
That figure is usually treated as weather (i.e. storms happen, costs follow.) Treated as a product specification instead, it highlights an operating model that can spot a problem hours or days earlier than it used to, and still cannot move until a person has opened a ticket, convened a call, and re-entered the same data into three systems.
Visibility platforms, control towers, risk scores, digital twins, and exception dashboards have defined the last decade of AI in the supply chain. That decade has been very good at collapsing the time between an event and awareness of it, but it has been far less good at collapsing the time between awareness and a commercial act.
Detection is a ‘solved-enough’ problem
Ask a chief supply chain officer where the AI budget went and the answer tends to follow a familiar list: demand sensing, ETA prediction, supplier risk scoring, inventory optimisation, and lane analytics. These tools work. Forecast error comes down. A vessel delay is flagged before the container misses the cut-off. A second-tier fab outage shows up on a heat map instead of in a customer email.
None of that accounts for the $184 billion. The bill is the interval after the flag: expedite or wait; split the order or accept the miss; retender the lane or pay the spot rate; consolidate two half-empty movements or ship both; swap ocean for air on the SKUs that actually justify the premium. These are bounded, repeatable decisions that sit inside policy, contract, and inventory limits the company already set—and they still queue behind a human inbox.
Surveys keep describing the same lag in different language. A 2026 Knosc survey of mid-market manufacturers and distributors found that supply-chain teams spend 28 percent of their working time responding to disruptions, most of it investigating what happened rather than changing what happens next.
Logistics executives still rank AI as a strategic priority (Capgemini’s 2025 research put an AI-driven “new-gen” supply chain among the top three technology trends for 70 percent of large-company executives) and then report that measurable financial impact remains rare. Gartner found in 2025 that only 23 percent of supply-chain organisations even have a formal AI strategy. The shortfall is not a shortage of models, but a shortage of authority granted to software.
The ticket is the product
Most current deployments are built around the ticket. The model produces a recommendation, the recommendation becomes an alert, the alert becomes a work item, and the work item waits for a planner already occupied with other work items. By the time the planner acts, the option set has narrowed—the alternative carrier’s capacity is gone, the consolidation window has closed, and the supplier’s next production slot is allocated.
That workflow is not a temporary step on the way to autonomy but the product companies bought. Vendors sold insight because insight is easy to demonstrate and easy to govern; action touches money, contracts, service levels, and blame. So the industry automated the part of the job that does not require a signature. FourKites and ABI Research reported in 2025 that only 27 percent of organisations allow AI to take autonomous action, while 52 percent confine it to decision support.
Adding another dashboard to a delayed shipment rarely moves EBITDA as a result. The decision cycle has not changed; it has only been decorated.
Bounded action as the next model
The firms set to take share are not the ones with the tidiest control tower but the ones that pre-authorise a narrow class of moves and let agents execute them while the exception is still cheap.
Retender a lane when the contracted carrier’s ETA slips beyond a threshold and a qualified alternate sits inside the approved rate band. Consolidate outbound waves when fill rates and cut-off times make a combined movement cheaper than two. Swap mode on a defined SKU set when the cost of air is lower than the cost of a missed retail window. Reallocate safety stock across two distribution centres when a forecast miss and a transport constraint line up.
None of that requires a strategy offsite. Each can be written as: if these conditions, then this action, within this spend cap, with this audit trail, and a human only if the case falls outside the fence. That is not a “lights-out” supply chain—it is the same discipline manufacturers already apply to machine control, where the agent may act inside the interlock and escalates outside it. The difference here is commercial rather than physical: the interlock is a policy object – category, supplier tier, mode, dollar limit, and service class – not a PLC.
Three conditions for real change
First, decisions have to be written as policies, not tribal knowledge. If the only place “we will pay air on A-items after 48 hours of ocean slip” lives is in a planner’s head, no agent can execute it. The work of the next two years is less model training than decision design: which moves are reversible, which are capped, and which suppliers and modes are pre-cleared.
Second, execution systems have to accept machine-initiated transactions. An agent that can draft an RFQ but cannot post it is still a detection tool. TMS, WMS, sourcing suites, and carrier APIs need to treat a bounded agent the way they treat a junior buyer with a spend limit—authenticated, logged, and reversible.
Third, accountability has to move with the action. If a retender inside policy goes wrong, the post-mortem should inspect the policy, the data, and the fence, not hunt for the person who “should have checked”. Until that cultural change happens, every agent will be designed to wait, because waiting is how careers survive.
The competitive split
For a while, both models will look alike on a slide—both will have AI, and both will have a control tower. The difference will show up in cycle time from detection to commercial act, and then in service and cost.
Companies that keep buying detection will know about the storm earlier. Companies that authorise bounded action will already have retendered the lane, consolidated the wave, and moved the A-items before the incident call is booked.
Disruption is not going away. Lead times in critical components, mode volatility, and multi-tier opacity are structural features of the network. What remains optional is whether the response waits for a human to open a queue. The product that created the lag was insight without authority. The product that ends it is an agent allowed to spend a little money, inside a fence, before anyone is free to look.
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.
Autonomous AI systems are beginning to move beyond software environments and into warehouses, delivery networks, and public spaces. The development is drawing attention to whether current AI rules cover systems that operate in physical environments.
Most existing AI governance frameworks have focused on online harms and model outputs, including bias, misinformation, and harmful content. Embodied AI systems carry risks in physical environments, where failures can affect infrastructure, propert
Autonomous AI systems are beginning to move beyond software environments and into warehouses, delivery networks, and public spaces. The development is drawing attention to whether current AI rules cover systems that operate in physical environments.
Most existing AI governance frameworks have focused on online harms and model outputs, including bias, misinformation, and harmful content. Embodied AI systems carry risks in physical environments, where failures can affect infrastructure, property, or human safety.
Singapore’s Infocomm Media Development Authority published version 1.5 of its Model AI Governance Framework for Agentic AI on May 20. The framework sets out guidance for organisations deploying AI agents that can plan, make decisions, and take actions across multiple steps to complete user-defined goals.
The framework says agents can interact with tools, external systems, and other agents, including systems that update databases, write files, control devices, or perform transactions. It lists access controls, monitoring, and human approval among governance measures for deployment.
AI moves into physical systems
At an AI summit in Singapore last week, discussions around robotics and embodied AI focused on operational safety issues more commonly associated with aviation, industrial systems, and critical infrastructure oversight than conventional software regulation.
Speakers also discussed whether autonomous systems can operate safely and reliably in unpredictable real-world environments over extended periods.
Dr. Ya-Qin Zhang, founding dean of the Institute for AI Industry Research at Tsinghua University, said embodied AI systems amplify risks already associated with autonomous software. He said failures can directly affect transport systems, drones, logistics networks, and critical infrastructure.
“Any risk in the digital domain will be amplified in the physical domain, and the physical domain will have a physical consequence,” Zhang told MLex on the sidelines of the summit.
He added that vehicles, drones, smart grids, and other infrastructure could become exposed as AI systems are embedded more deeply into physical operations.
Speakers discussed reliability, operational monitoring, and post-deployment assurance as governance concerns. Summit discussions pointed to deployment-based governance models built around simulation, telemetry, and iterative testing, rather than one-time certification alone.
IMDA’s framework also recommends gradual rollouts, continuous monitoring, and further testing after deployment. It says agents interact dynamically with their environment and not all risks can be anticipated before release.
Monitoring becomes a deployment issue
Grab, which is piloting autonomous vehicles and delivery robots in Singapore’s Punggol district, said deployment governance depends heavily on simulation, testing, and continuous monitoring.
“We do a lot of simulation, we do a lot of testing in closed courses and open courses in order to make sure our robots are reliable,” Suthen Thomas Paradatheth, Grab’s chief technology officer, said during one of the summit panels.
“Before we scale to hundreds of robots, we make sure we crack it first in simulation and with a few robots,” he added.
Grab also pointed to monitoring systems designed to track robot performance and detect unexpected failures after deployment.
“There’s a long tail of issues that could emerge,” Paradatheth said.
The IMDA framework says organisations should assess agentic AI use cases based on data access, external system access, autonomy, and task complexity. It also points to the scope and reversibility of agent actions, third-party involvement, and overall system complexity.
It also recommends limiting agent access to tools and systems, applying least-privilege permissions, and defining standard operating procedures for agent workflows. Organisations should also set mechanisms to take agents offline when they malfunction.
Accountability spreads across more actors
MLex reported that embodied AI systems can involve several parties across development, manufacturing, and deployment. These include AI developers, robotics manufacturers, semiconductor suppliers, and infrastructure operators.
MLex also noted that responsibility can be harder to assign when systems continue adapting after deployment through software updates, telemetry, and operational data.
IMDA says organisations and humans remain accountable for agent actions, even when agents operate autonomously. The framework calls for clear responsibility across the agentic AI value chain, from model and platform providers to deployers, tooling providers, and end users.
Applied Materials said large-scale robotics deployment is also tied to semiconductor economics and systems integration. Om Nalamasu, the company’s chief technology officer, said robotics systems will depend on better sensors, energy efficiency, advanced packaging, and computing architectures.
Nalamasu said robotics systems would require purpose-built designs adapted to specific industrial ecosystems rather than a single solution for all environments.
Zhao Yuli, chief strategy officer of Chinese robotics startup Galbot, said Beijing is prioritising deployment scale and industrial commercialisation through government-backed testbeds, industrial partnerships, and long-term funding initiatives.
Galbot has deployed humanoid robotics systems in retail, warehouse, and pharmaceutical operations in China. These include autonomous stores that operate around the clock. Zhao said semi-structured industrial environments are likely to become an early commercialisation path because they offer more controllable operating conditions.
Japan is placing more focus on standards-setting, robotics datasets, and safety governance. Professor Yutaka Matsuo of the University of Tokyo’s Graduate School of Engineering pointed to an “AI Association” project aimed at collecting 100,000 hours of robotics data to support robotic foundation models.
Matsuo also referred to Japan’s AI Safety Institute and the Hiroshima AI Process as part of broader efforts to develop governance standards for embodied AI systems with Singapore and other Asian countries.
Singapore sets out agent controls
Singapore’s framework sets out four governance areas for agentic AI. These cover upfront risk assessment, human accountability, technical controls, and end-user responsibility. The framework describes them as an iterative process rather than a one-time assessment.
The framework says human oversight has to be adapted for agentic systems because continuous review of all workflows becomes impractical at scale. It recommends human approval at significant checkpoints, including high-stakes actions, irreversible actions, and outlier behaviour.
IMDA also identifies automation bias and alert fatigue as risks when humans supervise capable agents. It recommends auditing oversight through indicators such as human override rates and response times, and using automated real-time monitoring to flag unexpected behaviour.
The framework says users should be told what actions an agent can take, what data it can access, and what responsibilities remain with the user. It also recommends employee training on human-agent interaction, oversight, and the professional skills needed to assess agent outputs.
Companies test AI in regulated workflows
JPMorgan is implementing AI tools across its global investment banking business, Paul Uren, the bank’s Asia Pacific head of investment banking, told Reuters. The bank said the tools help bankers access more information and synthesise it with internal systems. They are also being used to prepare content and support client engagement.
JPMorgan CEO Jamie Dimon told Bloomberg News that the bank would hire more AI specialists and fewer traditional bankers. Reuters reported that global banks are increasing AI investment, reshaping workforces, and changing job roles.
The bank is also among selected organisations permitted by Anthropic to use its Mythos cybersecurity model under a controlled initiative known as Project Glasswing. According to Anthropic, Mythos can detect old vulnerabilities in browsers, infrastructure, and software.
Reuters reported that Goldman Sachs, Citigroup, Bank of America, and Morgan Stanley also have access to, or are testing, Mythos, citing sources and company executives.
IMDA’s framework includes a case study from OCBC Bank of Singapore on source-of-wealth analysis. The system parses income-related documents and drafts a source-of-wealth memo. It does not make credit, onboarding, or risk decisions autonomously.
In that case, the workflow is limited to task-level autonomy and operates only when triggered by predefined workflows. Human review is required at critical decision points, and final validation remains with designated reviewers.
Robots move into industrial use
In Japan, one-third of companies are already using or considering AI-powered robots, according to a Reuters survey conducted by Nikkei Research from May 1 to May 15. The survey contacted 492 companies, with 220 responding on the condition of anonymity.
About 4% of respondents said they already use AI robots, 5% plan to deploy them, and 25% are considering doing so. The remaining 66% said they had no such plans.
Transportation equipment manufacturers were the most active group in the survey, with 80% already using AI robots or considering deployment. By comparison, 94% of wholesale sector respondents said they had no plans to deploy AI robots.
Among companies using, planning to use, or considering AI robots, 71% selected manufacturing as a use case. Another 19% selected dangerous tasks, while 11% selected customer-facing services.
The Japanese government expects AI robots to help address the country’s chronic labour shortage and support its position in industrial robotics. Japan is home to robotics companies including Fanuc, Yaskawa Electric, and Kawasaki Heavy Industries, but faces competition from China and the United States in AI-enabled robotics.
Retail agents expand beyond search
Walmart has outlined plans to use agentic AI across shopping, employee, supplier, and developer workflows.
In July 2025, the retailer announced plans for four AI-powered “super agents.” They are designed for shoppers, store employees, suppliers and sellers, and software developers. Walmart said these agents would become the main entry point for AI interactions across those groups.
One of the tools, Sparky, is already available in Walmart’s app as a generative AI-powered shopping assistant. Hari Vasudev, Walmart’s US chief technology officer, said its expanded version would be able to reorder items and plan events. It would also use computer vision to suggest recipes based on the contents of a shopper’s fridge.
Walmart is also developing an Associate super agent for store workers and corporate staff. A separate Marty agent is being built for sellers, suppliers, and advertisers. The retailer is also working on a Developer super agent for testing, building, and launching future AI tools.
The company declined to say whether the agents would replace jobs. Dave Glick, senior vice president of enterprise business systems, said the tools would create new jobs, without giving further details.
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.
OpenAI will open its first Applied AI Lab outside the US in Singapore. The lab is part of a new partnership with the Ministry of Digital Development and Information.
The initiative, called OpenAI for Singapore, was announced at the ATx Summit and is backed by a commitment of more than S$300 million.
The lab will create more than 200 Singapore-based technical roles over the next few years. OpenAI said Singapore will also become one of its global hubs for forward-deployed engineers who will work w
OpenAI will open its first Applied AI Lab outside the US in Singapore. The lab is part of a new partnership with the Ministry of Digital Development and Information.
The initiative, called OpenAI for Singapore, was announced at the ATx Summit and is backed by a commitment of more than S$300 million.
The lab will create more than 200 Singapore-based technical roles over the next few years. OpenAI said Singapore will also become one of its global hubs for forward-deployed engineers who will work with organisations on AI deployment. OpenAI said the lab’s work will be aligned with Singapore’s AI Mission priorities which include public service, finance, and digital infrastructure.
Focus on deployment and talent
The company will work with government agencies and local partners on education and workforce programmes within the Ministry of Education and GovTech. OpenAI also plans to support educators through a Singapore chapter of the OpenAI Academy, participate in the National AI Impact Programme, and run Codex for Teachers hackathons.
The partnership includes plans to work with local partners on accelerator programmes for AI-native startups in the form of workshops for micro-entrepreneurs and small businesses, covering how founders and SMEs can use AI in operations and customer service.
Chng Kai Fong, Permanent Secretary for Digital Development and Information, said Singapore’s response to AI includes growing new sectors, anchoring global frontier companies, and equipping workers with relevant skills.
Singapore updates agentic AI framework
Singapore has also updated its governance framework for agentic AI, which was launched by the Infocomm Media Development Authority at the World Economic Forum in January 2026. The framework builds on Singapore’s earlier Model AI Governance Framework for AI, introduced in 2020, and gives organisations guidance on the responsible deployment of AI agents, including measures to reduce the risks inherent in agentic AI.
IMDA has now updated the framework after seeking feedback and case studies from the industry, with the revised version following input from more than 60 organisations, including AWS, DBS, Google, and Salesforce.
The update adds guidance on risks linked to multi-agent systems, third-party agents, automation bias, and human accountability. The framework now includes more than ten case studies showing how organisations have applied its recommendations.
The case studies were contributed by Singaporean and international organisations, including Ant International, City Developments Limited, Cyber Sierra, Dayos, Google, Knovel, OCBC, PwC, Stability Solutions, Tencent, Terminal 3, Workday, X0PA, and GovTech Singapore.
Case studies show governance controls
One case study focuses on Dayos, a Singapore-headquartered enterprise AI automation company with operations in the US. Dayos built an AI-powered ticketing agent that handles internal IT requests. The agent can resolve some requests automatically and route requests to a human when needed.
Dayos used tiered risk levels to determine what actions the agent could take. Low-risk and reversible actions, like password resets, could be automated and audited biweekly, while moderate-risk actions required human approval before execution. Higher-risk actions, like permission changes with limited reversibility, were excluded from the agent’s authority.
Tencent contributed a case study on CodeBuddy, an agentic AI coding system developed by Tencent Cloud. CodeBuddy can plan, write, and deploy code through natural language instructions and can access filesystems, terminal commands, external APIs, and MCP tools.
CodeBuddy uses preset defaults and configurable permissions. Human approval is required for actions like editing files, running shell commands, making network requests, or using external tools.
The system explains complex commands in plain language before users approve them. Suspicious commands still require human approval, even if similar commands had been pre-approved.
GovTech Singapore’s case study covers the rollout of agentic coding assistants in government. The first phase was limited to GovTech employees, did not allow external tools, and was restricted to low-risk systems. GovTech developed central logging and a framework for connecting approved external tools. The agency also tested the system against potential attacks.
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.
Alibaba has unveiled a new AI processor built specifically for AI agents, pairing the chip announcement with a multi-year silicon roadmap and a new large language model, signalling that the company is building an integrated AI stack rather than just filling a gap left by US export controls.
The Zhenwu M890, developed by Alibaba’s semiconductor subsidiary T-Head, delivers three times the performance of its predecessor, the Zhenwu 810E, according to the company, as per Reuters report. But the p
Alibaba has unveiled a new AI processor built specifically for AI agents, pairing the chip announcement with a multi-year silicon roadmap and a new large language model, signalling that the company is building an integrated AI stack rather than just filling a gap left by US export controls.
The Zhenwu M890, developed by Alibaba’s semiconductor subsidiary T-Head, delivers three times the performance of its predecessor, the Zhenwu 810E, according to the company, as per Reuters report. But the performance jump is less notable than the architectural intent behind the chip: the M890 is purpose-built for AI agents, where software systems must retain long stretches of context, coordinate with other models in real time, and execute complex multi-step tasks with limited human intervention.
Those demands, heavy on memory bandwidth and inter-model communication, are meaningfully different from what standard inference chips are optimised for. The difference matters because it tells you something about where Alibaba thinks AI compute is heading. The company isn’t designing around today’s dominant use case; it’s building for the workload profile it expects to define enterprise AI over the next several years.
Built for AI agents, not just inference
More significant than the chip itself is the roadmap Alibaba put alongside it. The M890 will be followed by the V900 in the third quarter of 2027, expected to deliver another roughly threefold performance gain, followed by the J900 in the third quarter of 2028. That’s a deliberate, sustained cadence of in-house silicon upgrades that mirrors the kind of tick-tock product cycles Nvidia has used to maintain its lead in AI accelerators.
The parallel to Huawei is worth noting. Huawei laid out a similar chip roadmap for its Ascend line last year, and both announcements reflect the same underlying reality: Chinese technology companies have concluded that depending on foreign silicon, even in scenarios where export restrictions might ease, is a structural risk they cannot accept. The response has been to treat semiconductor development as a long-term capability-building exercise rather than a procurement problem.
Alibaba’s commitment to that exercise is not shallow. The company pledged more than 380 billion yuan, roughly US$53 billion, on cloud and AI infrastructure over three years last year, its largest-ever investment commitment to the sector. The M890 and its successors are downstream of that spending.
Traction that predates the announcement
T-Head said it has shipped more than 560,000 Zhenwu units to date, with over 400 external customers across 20 industries deploying the chips, including automakers and financial services firms. That is a material production footprint, not lab hardware, and it provides Alibaba with real-world deployment data at scale ahead of the M890’s rollout.
The new chip will be available to Chinese enterprise customers through Alibaba Cloud’s domestic model platform, Bailian, packaged inside the Panjiu AL128, a server system that stacks 128 M890 accelerators into a single rack.
The software side of the stack
Alongside the hardware, Alibaba announced Qwen 3.7-Max, the latest version of its flagship large language model, described as engineered for advanced coding and long-running agent tasks. The company said the model can operate continuously for up to 35 hours without performance degradation, a capability specification that only makes sense if you are designing for extended autonomous operation.
The timing is deliberate. Releasing a chip and a model optimised for the same workload class on the same day is a platform play. Alibaba is building a closed loop: its own silicon in T-Head, its own model in Qwen, its own cloud delivery in Bailian. Each component reinforces the others, and the combined stack is designed to reduce enterprise customers’ dependence on any external vendor.
More than half a million chips have been shipped. A successor is arriving in 2027, with another planned for 2028. T-Head is not hedging. At some point, building around US export controls stops being a workaround and starts being a strategy. Alibaba appears to have crossed that line.
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 AI and Big Data programme on day two of TechEx North America referred at least once to the “AI graveyard,” meaning the large number of pilots that never become durable systems. That phrase set the tone. The question was proof.
The Enterprise AI Implementation, ROI and Adoption track dealt with the hard middle of AI work. Its sessions covered stalled pilots, agentic AI for business impact, the move from experimentation to impact, the decision to buy or build, and durable ROI and autonomous de
The AI and Big Data programme on day two of TechEx North America referred at least once to the “AI graveyard,” meaning the large number of pilots that never become durable systems. That phrase set the tone. The question was proof.
The Enterprise AI Implementation, ROI and Adoption track dealt with the hard middle of AI work. Its sessions covered stalled pilots, agentic AI for business impact, the move from experimentation to impact, the decision to buy or build, and durable ROI and autonomous decisioning. A system has to be adopted, governed and measured before it deserved to be called successful.
The session on the AI graveyard was useful because it named the failure pattern. Many companies have enough budget to start AI experiments and enough executive attention to publicise them. Fewer have the data quality, process design, operating authority, and risk control to keep them going.
A day-two session on moving beyond copilots towards agentic AI framed the issue as business impact not novelty. Copilots have been useful as individual productivity tools, but their value is often hard to measure. Agents promise a closer connection to business process, yet they also increase the need for boundaries. An agent that can act in systems has to be evaluated by the quality of the action.
That point linked directly with the Future of AI track. Its opening theme, trust as a competitive advantage, was a useful counterweight to speed. The programme dealt with transparency, governance, regulation, banking analytics, and risk. It also included material from Hex on data agent, with evaluation and governance built in. Agentic AI will not mature in enterprise settings if evaluation remains informal.
Governance appeared in several forms. There was cross-functional governance, which reflects the reality that AI risk does not belong to legal, security or engineering. There was governance in the data layer, where trust depends on lineage and quality. There was governance around agent personas and risk stacks, where companies need to understand what an AI agent is permitted to know and do. The banking session gave the theme a sectoral focus, since financial services have less room for undefinedassurances about automation.
Digital Transformation Week carried the same day-two pressure into business delivery. The programme was built around real use cases, business impact, ROI, AI agents built on APIs, change readiness, government service transformation, city innovation and the conversion of data into financial value. The change-readiness material was especially important. AI fails because staff do not change routines, managers do not alter incentives, or the data needed for daily use never appears in the right place.
Sessions involving the DMV and the City of San Jose placed AI and transformation inside government service. In government, the measure of quality includes reliability, access, explainability and public trust. The Dow material on turning data into dollars sat at the commercial end of the same argument. In both cases, value depends on connecting data work with an accountable outcome.
The Cyber Security and Cloud Expo day-two programme expanded on risk. Its cloud-first enterprise track dealt with AI-led threats, cloud security, the “GenAI velocity gap,” threat intelligence, identity security and AI governance. The cyber programme treated AI as a force that changes attack and defence alike. It can help automate defensive work, but it can also accelerate misuse, widen leakage routes, and increase the strain on existing controls.
The phrase “velocity gap” was used several times during day two. Business units are adopting generative AI faster than many security teams can oversee it: the tools arrive first, policy and monitoring arrive later. The sessions on jailbreaking and data leaks made the point more concretely. If staff place sensitive material into unsanctioned tools, or if approved AI systems are poorly bounded, cloud security and data governance become one and the same.
Zero trust was presented as one answer, with a stronger interpretation of zero trust must now include AI systems, agents, and the data around them. Identity is not limited to human users, but services, agents and automated workflows require permission models as well. The cloud-first enterprise is therefore becoming a place where identity, data classification, AI governance, and threat detection are part of the same control mechanisms.
(Image source: TechEx/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.
AI agents hold the promise of automatically moving data between systems and triggering decisions, but in some cases, they can act without a clear record of what, when, and why they undertook their tasks.
That has the potential to create a governance problem, for which IT leaders are ultimately responsible. If an organisation can’t trace an agent’s actions and don’t have proper control over its authority, leaders can’t prove that a system is operating safely or even lawfully to regulators.
AI agents hold the promise of automatically moving data between systems and triggering decisions, but in some cases, they can act without a clear record of what, when, and why they undertook their tasks.
That has the potential to create a governance problem, for which IT leaders are ultimately responsible. If an organisation can’t trace an agent’s actions and don’t have proper control over its authority, leaders can’t prove that a system is operating safely or even lawfully to regulators.
That’s an issue set to become more important from August this year, as enforcement of the EU AI Act kicks in. According to the text of the Act, there will be substantial penalties for failures of governance relating to AI, especially when used in high-risk areas such as when personally-identifiable information is processed, or financial operations take place.
What IT leaders need to consider in the EU
Several steps can be taken to alleviate high levels of risk, and of these, the ones that stand out for consideration include agent identity, comprehensive logs, policy checks, human oversight, rapid revocation, the availability of documentation from vendors, and the formulation of evidence for presentation to regulators.
There are several options decision makers can consider that will help create the record of activities undertaken by agentic systems. For example, a Python SDK (software development kit), Asqav, can sign each agent’s action cryptographically and link all records to an immutable hash chain – the type of technique that’s more associated with blockchain technology. If someone or something changes or removes a record, verification of the chain fails.
For governance teams, using a verbose, centralised, possibly-encrypted system of record for all agentic AIs is a measure that provides data well beyond the scattered text logs produced by individual software platforms. Regardless of the technical details of how records are made and kept, IT leaders need to see exactly where, when, and how agentic instances are acting throughout the enterprise.
Many organisations fail at this first step in any recording of automated, AI-driven activity. It’s necessary to keep a registry of every agent in operation, with each uniquely identified, plus records of its capabilities and granted permissions. This ‘agentic asset list’ ties neatly into the requirements of the EU AI Act’s article 9, which states:
Article 9: For high-risk areas, AI risk management has to be an ongoing, evidence-based process built into every stage of deployment (development, preparation, production), and be under constant review.
Furthermore, decision-makers need to be aware of the Act’s Article 13:
High-risk AI systems have to be designed in such a way that those deploying them can understand a system’s output. Thus, an AI system from a third-party must be interpretable by its users (not an opaque code blob), and should be supplied with enough documentation to ensure its safe and lawful use.
This requirement means the choice of model and its methods of deployment are both technical and regulatory considerations.
Putting the brakes on
It’s important for any agentic deployment to offer a facility for the revocation of an AI’s operating role, preferably within a matter of seconds. The ability to revoke quickly should be part of emergency response processes. Revocation options should include the immediate removal of privileges, immediate ceasing of API access, and the flushing of queued tasks.
The presence of human oversight, combined with the presentation of enough context for humans to make informed decisions, means that human operators must be able to reject any proposed action. It’s not considered adequate for the person reviewing a decision to see only a prompt or a confidence score. Effective oversight needs information around context, every agent’s authority, and time enough to intervene to prevent mis-steps.
Multi-agent considerations
While every agent’s action should be recorded automatically and retained, multi-agent processes are particularly complex to track, as failures can take place among chains of agents. It’s therefore important for security policies to be tested during the development of any system that intends to utilise multiple agents.
Finally, governing authorities may require logs and technical documentation at any time, and will certainly need them after any incident they have been made aware of.
Conclusion
The question to be considered by IT leaders considering using AI on sensitive data or in high-risk environments is whether every aspect of the technology can be identified, constrained by policy, audited, interrupted, and explained. If the answer is unclear, governance is not yet in place.
(Image source: “Last Judgement” by Lawrence OP is licensed under CC BY-NC-ND 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-nd/2.0)
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.
Anthropic’s most capable AI model has already found thousands of AI cybersecurity vulnerabilities across every major operating system and web browser. The company’s response was not to release it, but to quietly hand it to the organisations responsible for keeping the internet running.
That model is Claude Mythos Preview, and the initiative is called Project Glasswing.
The launch partners include Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Fou
Anthropic’s most capable AI model has already found thousands of AI cybersecurity vulnerabilities across every major operating system and web browser. The company’s response was not to release it, but to quietly hand it to the organisations responsible for keeping the internet running.
That model is Claude Mythos Preview, and the initiative is called Project Glasswing.
The launch partners include Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, Nvidia, and Palo Alto Networks.
Beyond that core group, Anthropic has extended access to over 40 additional organisations that build or maintain critical software infrastructure. Anthropic is committing up to US$100 million in usage credits for Mythos Preview across the effort, along with US$4 million in direct donations to open-source security organisations.
A model that outgrew its own benchmarks
Mythos Preview was not specifically trained for cybersecurity work. Anthropic said the capabilities “emerged as a downstream consequence of general improvements in code, reasoning, and autonomy”, and that the same improvements making the model better at patching vulnerabilities also make it better at exploiting them.
That last part matters. Mythos Preview has improved to the extent that it mostly saturates existing security benchmarks, forcing Anthropic to shift its focus to novel real-world tasks–specifically, zero-day vulnerabilities. These flaws were previously unknown to the software’s developers.
Among the findings: a 27-year-old bug in OpenBSD, an operating system known for its strong security posture. In another case, the model fully autonomously identified and exploited a 17-year-old remote code execution vulnerability in FreeBSD–CVE-2026-4747–that allows an unauthenticated user anywhere on the internet to obtain complete control of a server running NFS. No human was involved in the discovery or exploitation after the initial prompt to find the bug.
Nicholas Carlini from Anthropic’s research team described the model’s ability to chain together vulnerabilities: “This model can create exploits out of three, four, or sometimes five vulnerabilities that in sequence give you some kind of very sophisticated end outcome. I’ve found more bugs in the last couple of weeks than I found in the rest of my life combined.”
Why is it not being released?
“We do not plan to make Claude Mythos Preview generally available due to its cybersecurity capabilities,” Newton Cheng, Frontier Red Team Cyber Lead at Anthropic, said. “Given the rate of AI progress, it will not be long before such capabilities proliferate, potentially beyond actors who are committed to deploying them safely. The fallout–for economies, public safety, and national security–could be severe.”
This is not hypothetical. Anthropic had previously disclosed what it described as the first documented case of a cyberattack largely executed by AI–a Chinese state-sponsored group that used AI agents to autonomously infiltrate roughly 30 global targets, with AI handling the majority of tactical operations independently.
The company has also privately briefed senior US government officials on Mythos Preview’s full capabilities. The intelligence community is now actively weighing how the model could reshape both offensive and defensive hacking operations.
The open-source problem
One dimension of Project Glasswing that goes beyond the headline coalition: open-source software. Jim Zemlin, CEO of the Linux Foundation, put it plainly: “In the past, security expertise has been a luxury reserved for organisations with large security teams. Open-source maintainers, whose software underpins much of the world’s critical infrastructure, have historically been left to figure out security on their own.”
Anthropic has donated US$2.5 million to Alpha-Omega and OpenSSF through the Linux Foundation, and US$1.5 million to the Apache Software Foundation–giving maintainers of critical open-source codebases access to AI cybersecurity vulnerability scanning at a scale that was previously out of reach.
What comes next
Anthropic says its eventual goal is to deploy Mythos-class models at scale, but only when new safeguards are in place. The company plans to launch new safeguards with an upcoming Claude Opus model first, allowing it to refine them with a model that does not pose the same level of risk as Mythos Preview.
The competitive picture is already shifting around it. When OpenAI released GPT-5.3-Codex in February, the company called it the first model it had classified as high-capability for cybersecurity tasks under its Preparedness Framework. Anthropic’s move with Glasswing signals that the frontier labs see controlled deployment–not open release–as the emerging standard for models at this capability level.
Whether that standard holds as these capabilities spread further is, at this point, an open question that no single initiative can answer.
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.
A survey carried out by OutSystems, The State of AI Development 2026 [email wall], argues that AI has moved into early production phase for many enterprises, primarily inside the IT function.
The survey was based on the responses of 1,879 IT leaders, and warns that adoption of AI is in danger of running ahead of governance and integration. The shortfall is a gap between what IT leaders want agents to do and what their organisations can safely control. The report’s authors urge companies to ad
A survey carried out by OutSystems, The State of AI Development 2026 [email wall], argues that AI has moved into early production phase for many enterprises, primarily inside the IT function.
The survey was based on the responses of 1,879 IT leaders, and warns that adoption of AI is in danger of running ahead of governance and integration. The shortfall is a gap between what IT leaders want agents to do and what their organisations can safely control. The report’s authors urge companies to address the controls or guardrails on AI systems, and also stress the importance of integrating new, AI technology into an organisation’s existing platforms.
OutSystems says 97% of its respondents are exploring some form of agentic strategy, with 49% of them describing their current abilities as “advanced” or “expert.” Nearly half of those surveyed say that over half of agentic AI projects have moved from pilot into production, with Indian companies most successful in implementing the technology: 50% of Indian companies say their AI projects are 51% to 75% successful.
Companies are considering where agents should be deployed first, and under what controls, but although “cost reduction or efficiency gains” is the most cited expectation for AI’s effects, only 22% found their deployments most effective in that regard. Instead, the most effective area gains in a business stemmed from equipping software developers with AI tools described as “generative AI-assisted.”
The report’s geography and sector data show that transitions to AI agentic workflows are unevenly distributed. India stands out as the market with the highest share of users considering themselves “expert”, while many organisations in Australia, Brazil, Germany, the Netherlands, the UK, and the US still identify as intermediate stage users. France and Germany are the most dubious of AI adoption, with Germany recording the highest share of leaders not using agentic AI in any form.
The sectors and functions invested in AI
Financial services and technology show the most movement from pilot to production, with many implementations in core business functions. The sector can be considered as having the most clear line of sight from automation to measurable returns in terms of income. The practical inference from the report’s findings would be for slower-moving sectors to copy the implementation workflows employed by the fintech industry: Start with narrow, high-volume workflows where performance can be measured and failures can be contained, and focus on the IT function.
According to the survey, generative AI-assisted development is now common in nine of the ten countries surveyed, alongside traditional coding, outsourced development, and SaaS customisation. It undercuts the notion that enterprises are moving into an AI-native or all-AI stack. In fact, most organisations add agents and AI-generated code on top of the processes already proven effective in their development environments.
Fragmented data no roadblock to AI progress
OutSystems finds that 48% of respondents see integration with legacy systems as the most important ability needed to expand agentic AI, and 38% say legacy systems are the main reason projects stall between pilot and production. Of the potential barriers to AI development that were offered as choices to the survey’s participants, more than 40% cited integration difficulties and legacy fragmentation the most problematic.
Organisations considering large data clean-up programmes (which many AI vendors advocate as a reason why deployments fail to reach production) may want to rethink, the report implies. The authors state agents can be built that can work well in complex data environments, as long as governance and integration are strengthened at the same time as AI implementation. Across the board, most sectors express “moderate trust” levels of agentic AI at around 50%, although responses from different business functions were not broken out in the survey results’ figures.
IT operations and software development
The financial returns are manifest mostly in IT functions themselves. The report says the most explored use cases are IT operations, at 55%, and data analysis, at 52%. Workflow automation follows at 36%, then customer experience at 33%. On realised return on investment, IT development and productivity lead by a margin, at 40%, ahead of operational efficiency at 22%. That distribution suggests that the first durable value from agentic AI is internal at developers’ desks rather than in customer-facing environments. Customer-facing deployments may still make sense, but the report indicates they require more trust in system performance, stronger controls, better orchestration, and an ability to create watertight oversight mechanisms.
Trust in and control of agents and governance
Trust in agentic AI, however, is improving. OutSystems reports that 73% of respondents express either high or moderate trust in letting agents to act autonomously, a rise of around 10% compared to a similar survey the company undertook last year. Trust in code or workflows generated by third-party AI tools is slightly lower, at 67%, a substantial increase from the prior year’s figure, when only 40% ‘mostly trusted’ generative AI to write code without human help.
Only 36% of respondents say they have a centralised approach to AI governance, while 64% say they lack such a facility, and 41% rely on rules implemented on a per-project basis. Two-thirds say building human-in-the-loop checkpoints is technically difficult because it requires orchestration that can pause agents – in effect inserting manual braking on operations that might be fully autonomous.
Many organisations appear to be deploying looser oversight models, although it is not clear if that is a result of greater trust in models or whether business functions are under pressure to deploy AI regardless of security or reliability concerns. If the trend to loosen oversight continues, the report’s authors note that agentic AI adoption may advance faster than the methods of accountability that many consider important.
Firms that want to scale agents in regulated or mission-critical settings should treat orchestration and auditability as part of the product, the survey’s findings state. When compliance checks consider a business’s operations, breadcrumb trails in the form of logfiles and defined responsibilities are considered important elements of any agentic AI rollout.
The report says 94% of leaders are concerned about “AI sprawl”, which is not defined, but could be inferred to be a lack of a centralised management platform that oversees all AI deployments in the enterprise. 39% are very or extremely concerned about the issue, and only 12% currently use a centralised platform to keep that sprawl under control.
(Image source: “Relax” by Koijots is licensed under CC BY-SA 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by-sa/2.0)
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.
A new open-source toolkit from Microsoft focuses on runtime security to force strict governance onto enterprise AI agents. The release tackles a growing anxiety: autonomous language models are now executing code and hitting corporate networks way faster than traditional policy controls can keep up.
AI integration used to mean conversational interfaces and advisory copilots. Those systems had read-only access to specific datasets, keeping humans strictly in the execution loop. Organisations ar
A new open-source toolkit from Microsoft focuses on runtime security to force strict governance onto enterprise AI agents. The release tackles a growing anxiety: autonomous language models are now executing code and hitting corporate networks way faster than traditional policy controls can keep up.
AI integration used to mean conversational interfaces and advisory copilots. Those systems had read-only access to specific datasets, keeping humans strictly in the execution loop. Organisations are currently deploying agentic frameworks that take independent action, wiring these models directly into internal application programming interfaces, cloud storage repositories, and continuous integration pipelines.
When an autonomous agent can read an email, decide to write a script, and push that script to a server, stricter governance is vital. Static code analysis and pre-deployment vulnerability scanning just can’t handle the non-deterministic nature of large language models. One prompt injection attack (or even a basic hallucination) could send an agent to overwrite a database or pull out customer records.
Microsoft’s new toolkit looks at runtime security instead, providing a way to monitor, evaluate, and block actions at the moment the model tries to execute them. It beats relying on prior training or static parameter checks.
Intercepting the tool-calling layer in real time
Looking at the mechanics of agentic tool calling shows how this works. When an enterprise AI agent has to step outside its core neural network to do something like query an inventory system, it generates a command to hit an external tool.
Microsoft’s framework drops a policy enforcement engine right between the language model and the broader corporate network. Every time the agent tries to trigger an outside function, the toolkit grabs the request and checks the intended action against a central set of governance rules. If the action breaks policy (e.g. an agent authorised only to read inventory data tries to fire off a purchase order) the toolkit blocks the API call and logs the event so a human can review it.
Security teams get a verifiable, auditable trail of every single autonomous decision. Developers also win here; they can build complex multi-agent systems without having to hardcode security protocols into every individual model prompt. Security policies get decoupled from the core application logic entirely and are managed at the infrastructure level.
Most legacy systems were never built to talk to non-deterministic software. An old mainframe database or a customised enterprise resource planning suite doesn’t have native defenses against a machine learning model shooting over malformed requests. Microsoft’s toolkit steps in as a protective translation layer. Even if an underlying language model gets compromised by external inputs; the system’s perimeter holds.
Security leaders might wonder why Microsoft decided to release this runtime toolkit under an open-source license. It comes down to how modern software supply chains actually work.
Developers are currently rushing to build autonomous workflows using a massive mix of open-source libraries, frameworks, and third-party models. If Microsoft locked this runtime security feature to its proprietary platforms, development teams would probably just bypass it for faster, unvetted workarounds to hit their deadlines.
Pushing the toolkit out openly means security and governance controls can fit into any technology stack. It doesn’t matter if an organisation runs local open-weight models, leans on competitors like Anthropic, or deploys hybrid architectures.
Setting up an open standard for AI agent security also lets the wider cybersecurity community chip in. Security vendors can stack commercial dashboards and incident response integrations on top of this open foundation, which speeds up the maturity of the whole ecosystem. For businesses, they avoid vendor lock-in but still get a universally scrutinised security baseline.
The next phase of enterprise AI governance
Enterprise governance doesn’t just stop at security; it hits financial and operational oversight too. Autonomous agents run in a continuous loop of reasoning and execution, burning API tokens at every step. Startups and enterprises are already seeing token costs explode when they deploy agentic systems.
Without runtime governance, an agent tasked with looking up a market trend might decide to hit an expensive proprietary database thousands of times before it finishes. Left alone, a badly configured agent caught in a recursive loop can rack up massive cloud computing bills in a few hours.
The runtime toolkit gives teams a way to slap hard limits on token consumption and API call frequency. By setting boundaries on exactly how many actions an agent can take within a specific timeframe, forecasting computing costs gets much easier. It also stops runaway processes from eating up system resources.
A runtime governance layer hands over the quantitative metrics and control mechanisms needed to meet compliance mandates. The days of just trusting model providers to filter out bad outputs are ending. System safety now falls on the infrastructure that actually executes the models’ decisions
Getting a mature governance program off the ground is going to demand tight collaboration between development operations, legal, and security teams. Language models are only scaling up in capability, and the organisations putting strict runtime controls in place today are the only ones who will be equipped to handle the autonomous workflows of tomorrow.
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.
AI systems are starting to move beyond simple responses. In many organisations, AI agents are now being tested to plan tasks, make decisions, and carry out actions with limited human input. It is no longer just about whether a model gives the right answer. It is about what happens when that model is allowed to act.
Autonomous systems need clear boundaries. They need rules that define what they can access, what they are allowed to do, and how their actions are tracked. Without those controls,
AI systems are starting to move beyond simple responses. In many organisations, AI agents are now being tested to plan tasks, make decisions, and carry out actions with limited human input. It is no longer just about whether a model gives the right answer. It is about what happens when that model is allowed to act.
Autonomous systems need clear boundaries. They need rules that define what they can access, what they are allowed to do, and how their actions are tracked. Without those controls, even well-trained systems can create problems that are hard to detect or reverse.
One company working on this problem is Deloitte. The firm has been developing governance frameworks and advisory approaches to help organisations manage AI systems.
From tools to AI agents
Most AI systems in use today still depend on human prompts. They generate text, analyse data, or make predictions, but a person usually decides what happens next. Agentic AI changes that pattern. These systems can break down a goal into steps, choose actions, and interact with other systems to complete tasks.
That added independence brings new challenges. When a system acts on its own, it may take paths that were not fully expected or use data in ways that were not intended.
Deloitte’s work focuses on helping organisations prepare for these risks. Rather than treating AI as a standalone tool, the firm looks at how it fits into business processes, including how decisions are made and how data flows through systems.
Building governance into the lifecycle
Governance should not be added after deployment. It needs to be built into the full lifecycle of an AI system.
This starts at the design stage. Organisations need to define what a system is allowed to do and where its limits are. This may include setting rules around data use and outlining how the system should respond in uncertain situations.
The next stage is deployment. At this point, governance focuses on access and control, including who can use the system and what it can connect to. Once the system is live, monitoring becomes the main concern. Autonomous systems can change over time as they interact with new data. Without regular checks, they may drift away from their original purpose.
The role of transparency and accountability
As AI systems take on more responsibility, it becomes more difficult to trace how decisions are made. This creates a demand for stronger transparency. Deloitte’s work highlights the importance of keeping track of how systems operate. This includes logging actions and documenting decisions. These records help organisations in determining what happened if something goes wrong. If an autonomous system takes an action, there needs to be clarity about who is responsible.
Research from Deloitte shows that adoption of AI agents is moving faster than the controls needed to manage them. Around 23% of companies already use them, and that figure is expected to reach 74% within two years. Only 21% report having strong safeguards in place to oversee how they behave.
Real-time oversight for AI agents
Once an autonomous system is active, the focus shifts to how it behaves in real-world conditions. Static rules are not always enough, and systems need to be observed as they operate.
Deloitte’s approach includes real-time monitoring, allowing organisations to track what an AI system is doing as it performs tasks. If the system behaves in an unexpected way, teams can step in quickly. This may involve pausing certain actions or adjusting permissions. Real-time oversight also helps with compliance. In regulated industries, companies need to show that systems follow rules and standards.
In practice, these controls are starting to appear in operational settings. Deloitte describes scenarios where AI systems monitor equipment performance across sites. Sensor data can signal early signs of failure, which can trigger maintenance workflows and update internal systems. Governance frameworks define what actions the system can take, when human approval is required, and how decisions are recorded. The process runs across multiple systems, but from a user’s point of view, it appears as a single action.
Governance is part of discussions at AI & Big Data Expo North America 2026, taking place on May 18–19 in Santa Clara, California. Deloitte is listed as a Diamond Sponsor for the event, placing it among the firms contributing to conversations around how autonomous systems are deployed and controlled in practice.
The challenge is not just building smarter systems, but ensuring they behave in ways organisations can understand, manage, and trust over time.
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.
Login
