The UK’s digital watchdogs have warned enterprises that they can’t evade their legal responsibilities by blaming errors on AI agents that they have deployed.

“By taking action on behalf of users, such as making purchases and communicating with businesses, these agents could transform how people use digital services, engage with markets and unlock wider innovation,” said the Digital Regulation Cooperation Forum (DRCF), a body comprising the UK’s Competition and Markets Authority (CMA), Financial Conduct Authority (FCA), Information Commissioner’s Office (ICO), and communications regulator Ofcom, in a joint statement.

While the DRCF welcomed such uses it advised, “firms developing or deploying personal agents should think carefully about how consumers can understand when and how an agent acts on their behalf, how authority to act is given and how their personal data is used.”

The DRCF didn’t pull its punches. “We want to be explicit that existing legal and regulatory frameworks apply to personal AI agents, including competition, consumer and data protection law, online safety, financial services, payments and cyber security requirements,” it said, adding, “An AI agent’s autonomy does not diminish the responsibility of organisations that deploy it to comply with applicable law and relevant regulation.”

There is certainly a degree of uncertainty over the legal position of autonomous AI agents. Gartner has warned that companies are increasingly facing legal challenges over “AI-induced” decisions, while attorneys have weighed in over gray areas in AI governance.

The four UK regulators aren’t immune to that doubt: Only a few weeks earlier, they issued a Call for Input on their 2027-27 work plan to help them avoid overlap and uncertainty about the interplay between their respective regulatory regimes.

The new year will see a new CEO for HP, which after a nine-month search has named Emma Walmsley to lead the company from January 4, 2027.

Walmsley was previously CEO of drug company GSK, where she turned around its finances and spun off its consumer healthcare business into a new company, Haleon. She has a reputation for courting confrontation rather than yes-men in the boardroom. HP needs a shake-up as revenue from printing declines and AI-driven demand for components is forcing up PC prices.

She will take over from HP’s interim CEO Bruce Broussard, who stepped in in February when former CEO Enrique Lores left the company after more than 25 years to become president and CEO at PayPal. There’s a consolation prize for Broussard, who will become chair of HP’s board, succeeding Chip Bergh who remains a board member.

Although Walmsley has been on Microsoft’s board for seven years (she is stepping down in December), this is her first foray into leading a tech company. Her almost 16-year tenure at GSK included more than eight years as CEO, and earlier in her career, she spent 17 years at L’Oréal.

Bergh said that during the executive search, she stood out “for the strength of her leadership experience across multiple industries, the clarity of her strategic thinking and her ability to translate customer insights into new growth.” He particularly cited her people-first leadership approach.

Walmsley reiterated her commitment to the philosophy in a post on LinkedIn. “HP has an extraordinary innovation heritage, an inspiring purpose and an incredible opportunity ahead as AI reshapes technology and what customers need from it. I’m excited to build on those strengths and help translate innovation into impact,” she wrote. “Behind every innovation are the people who work to make someone’s day a little easier, their work more productive, their ideas possible. In an age of AI, I believe human leadership has never mattered more. It’s a privilege to be joining this team.”

Announcing Walmsley’s appointment Broussard said, “We need a leader who can build on our innovation heritage and strengths to boldly shape the company’s future. Emma is a proven CEO with a track record of leading organizations at scale and guiding them through significant change. She has the experience and skills to bring HP’s businesses, capabilities, and people together to capture the opportunities ahead as AI moves to the edge, transforming how we work and live in a more connected world.”

This article first appeared on Computerworld.

Over many years of working through successive waves of enterprise technology and IT architecture, including fast-moving AI, I’ve heard the same question in different forms: “How can I build my business with AI while avoiding lock-in and tech debt?”

It came up again during a recent conversation about a major technology investment, when a CIO asked, “Should we really be making this commitment if the technology is changing constantly?” That resonated with me because it reflects a pattern I have seen across technology waves: the architecture must provide stability even as the capabilities around it continue to evolve.

For decades, the goal has been to build a technology foundation that serves the enterprise for five, seven or even 10 years. With the emergence of AI, some technologies become outdated before implementation is complete.

If shorter technology lifecycles are the challenge, adaptability is the answer. Instead of asking, “What will serve us for the next decade?” CIOs should ask, “How can we remain flexible as technology changes?” That requires being deliberate about what must remain durable and where the enterprise must be ready to adapt.

Think of it like a house. The foundation, walls and furnace are built to last, while paint, furniture and bathrooms change as needs and styles evolve. Enterprise architecture increasingly works the same way.

AI is separating the experience of work from the systems behind it

I recently heard about a company that now gives new employees a single onboarding tool: ChatGPT, connected to the enterprise systems and key data sources behind it. Instead of learning multiple applications, employees simply ask for the information or support they need, without knowing which underlying system provides it.

Of course, a single interface doesn’t mean blanket access to enterprise systems. This model requires security measures that ensure employees can access only the information and capabilities they’re authorized to use: carrying each employee’s identity and existing permissions through every request, limiting access to what’s necessary, tracking where information comes from, and logging retrievals and actions. Changes with significant business consequences should also require review.

This does change how we think about work and the technology that supports it. If agents are becoming the primary interface, the central question shifts from “Which application should employees use?” to “Can our data be accessed, understood and used securely by the interfaces and agents we adopt next?”

Adaptability means running at two speeds

Some technology capabilities should endure; others should evolve quickly. The mistake is treating everything as having the same lifecycle.

The durable layer includes ERP, HCM and other core systems, along with data, security and foundational architecture. These investments remain essential, but CIOs should design them around future business possibilities rather than solely around today’s processes.

The fast-moving layer comprises agents, interfaces, analytics, workflows, AI models and tools, experiments and business-specific solutions. The CIO’s role is to distinguish this layer from the durable core and ensure that rapidly evolving capabilities can be replaced without destabilizing the durable foundation.

CIOs need practical ways to change the underlying technology without disrupting the applications the business relies on. That starts with clear, consistent rules for how systems exchange data and interact with APIs, MCP connections and AI agents, keeping the implementation choices behind those boundaries. Before changing a model, agent or application, teams should confirm existing data and integrations will keep working, test security and expected behavior, and introduce changes in stages with a rollback plan if something goes wrong.

This two-speed model is a substantial departure from traditional enterprise IT management and from the way we did things when I was coming up in the tech world. It changes how CIOs plan investments, organize technology functions, structure vendor relationships and govern the connections between durable and fast-moving capabilities.

Invest thoughtfully for both durability and adaptability

In the pre-AI era, CIOs prioritized technologies with decade-long relevance. Today, they should avoid overspending on solutions prone to rapid obsolescence and invest in business outcomes and a durable foundation for intelligence.

Projects lasting more than six months warrant scrutiny: Will the solution remain relevant, or are we embedding processes that AI may soon handle; perhaps with a single prompt?

Durability extends beyond core systems to include the organization’s capacity to understand, evaluate and adopt new technology.

Today, most companies critically underinvest in skills. Knowing how to operate ChatGPT or Claude is not the same as working effectively with AI. OpenAI’s study found that AI power users outperform average users by sixfold. Upskilling gives CIOs leverage by helping employees redesign work and use AI effectively, while stronger leadership fluency facilitates new business models rather than merely supporting existing operations.

Most companies are also underinvesting in the data foundation. In West Monroe’s recent survey of 417 business leaders, AI governance, risk and compliance ranked as the lowest investment priority for the next 12 months. Traditionally, IT addressed data operations to ensure successful ERP implementation. Now, because data quality directly impacts AI value, investing in governance and accessibility is imperative.

CIOs need practices that preserve data meaning and reliability as platforms change. Those include assigning clear ownership, establishing authoritative sources and shared definitions, tracking data lineage and setting standards for quality and access. Definitions and rules should be maintained in portable documentation and data contracts, rather than embedding them in a single AI model or vendor platform, so they carry forward as technology evolves.

Break down IT silos to encourage adaptive architecture

Typically, CIOs organize operations around disciplines such as infrastructure, data, applications and security, with IT teams choosing an application and then defining processes and architecture around it. But as future architecture becomes less certain, CIOs must instead envision future business capabilities and design for change, bringing AI and data into architectural thinking alongside traditional cloud and application expertise.

Consider a reporting process that today moves from data analysis to Excel to PowerPoint to human review before a decision is made. An adaptive architecture focuses less on preserving each of those steps and more on the outcome: turning data into intelligence that informs a decision or action. Excel and PowerPoint may still have a role, but they no longer need to be hardwired into the process.

CIOs also need to break down silos. Teams often define architecture around a single suite, such as Salesforce or SAP, without considering alternatives. While working with an energy company recently, I saw how bringing analytics teams into the discussion shifted the focus from the ERP itself to how data should be accessed and used. Small, cross-functional teams can help create that broader view and design around business needs rather than individual platforms.

Build partnerships around future business outcomes

The most effective CIOs I work with are shifting their focus from simply controlling costs to driving value, actively partnering with the business, and reimagining how functions can evolve with AI, data and IT.

As vendor offerings expand across applications, analytics, agents and interfaces, CIOs need to balance integrated ecosystems with the flexibility to combine providers. Prioritizing open standards, multi-cloud support, data portability, interoperability and adaptable contracts can help organizations use the strongest elements of core systems while integrating agents and data across systems to adapt faster as needs evolve.

CIOs should build flexibility into vendor contracts, so changing providers doesn’t mean losing access to valuable data, context or governance rules. Contracts should provide for usable exports of data and metadata, documented interfaces that vendors keep current, advance notice of discontinued features, and support for testing portability and transitioning to another provider. They should also specify security and audit obligations throughout the relationship and during the transition.

The bottom line: buy with the future in mind. Don’t just ask if an application meets current requirements. Instead, envision what future business functions should look like, then derive technology needs from that vision. Shift the mindset from using HCM checklists to driving workforce optimization, from CRM checklists to revenue management, and from ERP checklists to enterprise data and intelligence.

Find the signal in the noise

As AI gets smarter and cheaper, the challenge will shift from affording it to connecting it to business outcomes. I believe the defining constraint in 2027 will be focus, not cost: CIOs must identify the few high-leverage priorities their organizations can absorb and execute well. That’s easier said than done.

There’s no established playbook for the AI era: we’re all experiencing it for the first time. CIOs face a flood of startups, models, architectural approaches, evolving standards, vendor claims and internal requests. The task becomes curation: deciding what deserves attention. Distinguishing signal from noise is becoming an essential leadership skill.

The new long game for CIOs

I believe CIOs are at a pivotal moment. They can remain excellent IT leaders or help the business rethink work through intelligence. In that sense, the role could evolve into what I call the chief intelligence officer: the leader who helps the enterprise turn fast-changing technology into durable business value.

Artificial intelligence has been part of semiconductor design for years, helping engineers optimize chip architectures for performance, power efficiency, silicon area and faster development. More recently, the industry has begun integrating large language models and AI agents into established electronic design automation (EDA) workflows. Now Cadence Design Systems is taking that approach further with its Maestro AI Model Portfolio, a collection of domain-specialized models designed to work with the company’s engineering software and AI Super Agents.

Maestro combines specialized model training, reinforcement learning and established EDA tools to improve how AI agents execute engineering tasks and evaluate results. Cadence claims better design outcomes with greater token efficiency versus general-purpose foundation models. While initially focused on chip and electronic system design, I also see broader implications for enterprise AI economics, proprietary data protection and deployment across complex engineering and business applications.

AI builds on established engineering software

Modern semiconductor design is enormously complex. Engineers must balance performance, power consumption, silicon area, manufacturing constraints and costs across chips containing billions of transistors. Cadence and chief competitor Synopsys have spent decades developing software to optimize for these challenges, and both are now incorporating AI agents capable of executing increasingly complex engineering workflows using those tools.

Cadence Design Systems, Inc.

“Revolutionizing complex engineering tasks requires deep domain-specialized intelligence,” notes Cadence president and CEO Anirudh Devgan in the company’s announcement. In my discussions with Cadence executives, we explored how Maestro builds on that approach to improve engineering workflows and design outcomes.

Cadence Maestro combines domain-specialized, post-trained language models, trained decision models and reusable agent skills with Cadence’s EDA and system design automation (SDA) tools. The portfolio supports ChipStack, InnoStack, ViraStack and AuraStack AI Super Agents, which interpret engineering objectives, break them into tasks and engage appropriate models and software tools. Rather than relying on a single model, agents can combine Maestro with commercial frontier and open-weight models according to workload requirements.

What I find particularly interesting is Cadence’s reinforcement learning architecture. EDA tools, evaluation frameworks and deterministic solvers assess AI-generated outputs against engineering constraints, including timing, power and physical design requirements. That feedback can improve subsequent model decisions and agent behavior. It’s a potentially important advantage over relying exclusively on general-purpose LLMs, since engineering results can be evaluated against measurable design constraints rather than language-based assessments alone.

NVIDIA is contributing its Nemotron model technology and collaborating with Cadence on agent security through its OpenShell AI sandboxing architecture, with integration work by both companies underway.

The business case for domain-trained AI in engineering

Cadence Design Systems, Inc.

Cadence reports that initial Maestro evaluations delivered up to 30% better quality of results and five times greater token efficiency, compared to a general-purpose foundation model. These are early vendor-reported results, however, and Cadence hasn’t disclosed enough benchmark detail to establish how consistently those advantages translate across production workloads.

Greater token efficiency is always a goal of course, particularly for AI agents executing lengthy, iterative engineering tasks. But reducing token consumption doesn’t always translate into proportional cost reductions. Engineering software, simulation infrastructure, verification and available compute resources also contribute substantially to development costs.

To me, the larger opportunity is improving design quality and shortening development cycles. Chip development programs can require hundreds of engineers and substantial capital investment, so even modest productivity improvements or faster time to market can offer significant value.

Cadence Design Systems, Inc.

Cadence customer Socionext expects up to fivefold productivity gains from deploying the company’s InnoStack AI Super Agent in production, although that’s a projection rather than a demonstrated Maestro result. Silicon Labs has also reported faster root-cause analysis and debugging using Cadence’s ChipStack technology. These examples show progress with agent-based engineering, but Maestro’s performance claims still need production validation. I’ll be interested to see how those results develop.

Deployment flexibility and data control

Another noteworthy aspect of Maestro is deployment flexibility. Cadence says its agentic engineering environment supports deployment through its own secure cloud platform, supported public clouds or customer-controlled on-prem infrastructure. Customers can also further train supported models using proprietary engineering data.

For semiconductor companies, where intellectual property can represent billions of dollars in investment, maintaining control over sensitive design information is critical. Similar considerations apply in financial services, healthcare and aerospace, where proprietary data and regulatory requirements can complicate AI deployment.

The tradeoffs warrant examination, however. On-prem deployments can shift responsibility for GPU infrastructure, model updates, security controls and optimization to the customer. Those costs need to be weighed against cloud inference expenses, data residency requirements and the potential performance benefits of processing data close to engineering workloads. In my view, Cadence’s deployment flexibility is valuable because enterprises can evaluate these choices against their infrastructure, security policies and engineering economics.

Specialized AI continues to gain momentum in chip design

Cadence Maestro is part of a broader industry push toward AI-driven semiconductor engineering. Established EDA vendors are expanding their agentic capabilities, while AI-native startups are developing systems for RTL generation, verification, debugging and physical design optimization. Domain-trained models, engineering feedback loops and intelligent model routing are also gaining attention.

What’s interesting here is the potential for agents to coordinate multiple engineering tools and models while retaining the design context needed for complex tasks. The business opportunity extends beyond automating individual steps to reducing the effort required to reach verified design results. For enterprise technology execs, that puts greater emphasis on toolchain integration, measurable productivity improvements and the economics of AI-assisted engineering.

Proving the value of specialized AI in chip design

Cadence’s first Maestro models are entering select customer early access in the fourth quarter of 2026, so it’s too early to know whether the company’s performance claims will translate consistently into production.

General-purpose models will remain important for broad reasoning, language and multimodal applications. However, specialized engineering processes often demand deeper domain knowledge, validated results and tight integration with existing software.

From my perspective, Cadence’s approach offers a glimpse of where enterprise AI is heading. Selecting the right models, tools and deployment architecture for specific workloads will be critical as adoption expands. And the early results are promising, with specialized AI showing the potential to improve engineering outcomes, boost productivity and reduce development costs at scale.

CIOs are dealing with a lot of change, from the rapid rise of AI and cloud modernization, to cybersecurity and workforce transformation. Many are sharing what they’re learning as they tackle these challenges, including what’s working, what isn’t, and how they’re implementing new technologies in their organizations. Here, a select group of tech leaders reveal the CIOs they follow and why.

Suma Nallapati, chief AI and information officer, City and County of Denver

I follow Suma Nallapati primarily through public technology and AI leadership discussions. She regularly discusses AI adoption, responsible AI, digital transformation, technology strategy, AI governance, human-centered technology, and using AI to improve public services. I find her perspective particularly valuable because she connects AI strategy with practical organizational and citizen outcomes. She also emphasizes responsible and human-centered AI rather than viewing AI simply as a technology implementation exercise. One initiative I found particularly interesting was Denver’s use of AI through its multilingual Sunny chatbot and other city services. Her current role includes scaling AI across areas such as resident engagement, internal automation, permitting, and emergency response while maintaining governance, transparency, and public trust. – Sachin Gupta, senior IT manager, Arrow Electronics

Jennifer Hartsock, chief information and digital officer, Cargill

I follow Jennifer Hartsock through the CIO community, as well as through her thought leadership, interviews, and LinkedIn. I especially enjoy hearing her perspective in longer-form conversations where she talks candidly about the realities of leading technology and transformation in a large, complex organization. Hartsock talks about many of the issues CIOs are navigating right now, including AI, digital transformation, technology operating models, talent, and leadership. A consistent theme in her perspective is connecting tech decisions back to the business, including how tech teams work with the business, and how you create greater ownership closer to the outcomes you’re trying to achieve. – Kathy Kay, EVP and CIO, Principal Financial Group

Fletcher Previn, SVP and CIO, Cisco

Fletcher Previn’s profile shows how a modern technology leader can discuss change without reducing it to a narrative about vendors. He explores AI, culture, talent, and productivity, with an emphasis on conditions that determine whether transformation lasts. The insight worth returning to is his focus on adoption as a leadership responsibility. New technology fails when teams lack learning time, managers lack clarity, and incentives reward behavior that no longer serves the business. CIOs can apply that lens to modernization, security initiatives, and data governance. It encourages an honest definition of progress based on decisions and outcomes rather than launch dates or dashboards. – Sherif Koussa, CEO, Software Secured

Kellie Romack, CDIO, ServiceNow

What makes Kellie Romack’s perspective valuable is she talks about AI from inside a large enterprise IT organization, not the outside. Her posts regularly cover enterprise AI, AI governance, autonomous IT, workforce transformation, cybersecurity, digital transformation, and how the CIO role is changing.

The point that stayed with me was her argument that CIO organizations now have to redesign jobs, teams, and workflows around AI while building new human skills at the same time. That’s a useful distinction. It frames successful AI adoption as an organizational and leadership problem as much as a technology one. Her seat at ServiceNow also gives her a view of what happens after the experiments stop. ServiceNow has reported that its internal AI use cases grew from 15 to more than 200. That makes her worth reading about for any CIO trying to understand what scaling enterprise AI actually looks like. – Tapabrata Biswas, project manager, InfluxIQ Tech Pvt. Ltd.

Mike Blandina, CIO, Snowflake

I follow Mike Blandina for his practical perspective on enterprise AI adoption and the operational realities behind digital transformation. His commentary consistently connects emerging technology with the governance, workforce skills, and measurable outcomes needed to make it useful. One insight that particularly resonates with me comes from this CIO.com article.

He makes the case for building an editing culture rather than a writing culture. As AI increasingly handles first drafts of code, reports, and other work, he argues that the critical human skill becomes evaluating outputs, identifying mistakes, and knowing when to challenge the system. I also appreciate his emphasis on data lineage, access permissions, and measuring actual productivity improvements instead of AI tool logins. Those are the less visible decisions that determine whether enterprise AI adoption delivers lasting value. – Parth Patel, director of operations, Shaligram

Deepa Soni, EVP and CIO, New York Life Insurance Company

Deepa Soni’s insights are particularly valuable because she oversees AI in one of the most critical and regulated sectors of financial services, a Fortune 100 life insurance company where AI decisions have real consequences for policyholders and regulators.

Soni frequently shares in-depth thoughts on what responsible AI adoption truly looks like within a large financial institution. She discusses the delicate balance between the speed of innovation and the readiness of governance, as well as how tech leaders can foster cultures that promote both agility and accountability. What sets her apart is that her approach to AI strategy is rooted in operational realities rather than just vendor hype, which is a rare and precious perspective for CIOs facing similar challenges. – Viren Meghani, technical architect and independent researcher in AI governance

Sears Merritt, CIO and head of enterprise technology and experience, MassMutual

I follow Sears Merritt on LinkedIn and through his interviews. He talks about what enterprise AI costs to run and how to keep the option to change your mind. Merritt also discusses why AI vendor contracts should be capped at 12 months to maintain flexibility, how success metrics are defined before AI pilots begin to determine whether they should be scaled, and the data platform he spent three years building before the AI wave arrived. This is useful to other CIOs because he’s one of the few CIOs at a regulated insurer who talks about AI in the language of procurement and measurement rather than vision. – Egiziago Cioffi, IT/enterprise architect and CEO, SynSphere Italia

Ganesh Ariyur, CIO and head of enterprise technology, Gainwell Technologies

I follow Ganesh Ariyur for his insightful views on AI, enterprise architecture, transformation, tech strategy, governance, and ways CIOs can link tech investments to quantifiable business value. What I find particularly helpful is his content extends beyond just talking about new technologies. He continually positions technology choices in relation to business results, operational frameworks, financial discipline, and implementation. This is particularly important for CIOs facing pressure to showcase ROI from AI and modernization efforts.

An example is his work, The 2026 CIO Operating Model: Architecture, AI, and Enterprise Value. His main point is that tech strategy must be viewed more like capital strategy, with CIOs assessing issues like if cloud expenditures enhance unit economics, and if architecture facilitates quicker decision-making. That viewpoint is important as it transitions the dialogue from which technology to implement, to what ways technology should transform the enterprise’s operations and value creation. – Ketan Patel, digital marketing manager, WeDoWebApps

Jim Swanson, EVP and CIO, Johnson & Johnson

I’m interested in how Jim Swanson thinks about the relationship between technology and the business, particularly where innovation carries considerable responsibility. For example, he recently discussed how AI agents are changing the way people interact with enterprise software. What caught my attention was his emphasis on the complexity that comes with that change — the integration, oversight, and accountability required to make it work. Those are questions that deserve as much attention as the technology itself. – Bill Huber, partner, ISG

Kimberly Basile, CIO, Kyndryl

I follow Kim Basile along with her bylined pieces on Kyndryl’s newsroom and interviews in the business press. She discusses enterprise AI adoption, workforce readiness, AI governance, platform modernization, and security leadership, informed by her earlier role as SVP and CISO at Leidos. Basile speaks from inside a transformation few CIOs have experienced, having helped guide Kyndryl’s technology separation from IBM and consolidate its application portfolio. She treats governance and people readiness as part of AI adoption from the start, and she’s candid about what hasn’t worked. In a recent feature on agentic AI, for instance, she said Kyndryl initially focused on counting agents, and the number didn’t tell the true story. Her team pivoted to measure business process flows. She also identified automating a process that hadn’t been properly validated as one of the company’s biggest mistakes. Activity metrics are easy to report, so organizations default to them. Her lesson pushes CIOs to validate a process before automating it.

She matters beyond the technology because women remain underrepresented in IT leadership, and she uses her visibility to lift others. When she was named to Women We Admire’s Top Women CIOs of 2026, her second year on the list, she used her post to congratulate the other honorees by name and highlight how these leaders help employees embrace technology. That generosity matters in a field where women CIOs are still too few. As the co-founder of Girl Power Talk, I pay attention to leaders who open doors for the next generation of women in tech. Basile does that consistently. – Sameer Somal, founder and CEO, Blue Ocean Global Technology

Enterprise collaboration technology has become an essential part of daily business operations, but managing it is increasingly complicated. What was once a relatively isolated audiovisual (AV) environment is now deeply integrated with enterprise IT infrastructure, unified communications and collaboration (UCC) platforms, and security systems.

For IT leaders, that convergence is creating new demands on resources, expertise, and budgets.

“Complexity has increased dramatically in the environments that customers have to manage today,” says Karen Klosinski, VP Product & Solutions at AVI-SPL.

Supporting these environments now requires expertise across multiple disciplines, while frequent software updates introduce a pace of change that many internal IT teams struggle to maintain. As collaboration environments become more complex, organizations are reconsidering whether traditional support models can keep pace.

Recognizing when IT needs additional support

Several indicators can signal that an organization’s approach to collaboration management needs to evolve. Platform migrations, for example, can require extensive retraining or hiring to develop new technical capabilities. IT leaders may also find that maintaining collaboration systems consumes resources needed for more strategic initiatives. Meanwhile, recurring technical issues and declining employee satisfaction can suggest that existing support models are falling short.

These challenges don’t necessarily require organizations to outsource everything. Managed services can supplement internal capabilities, giving IT teams access to specialized expertise without the expense and difficulty of building every competency in-house.

That economic argument is becoming increasingly relevant to smaller organizations. As demand for specialized skills grows, even midsize businesses may struggle to justify the cost of hiring experienced administrators for platforms such as Microsoft Teams.

Moving beyond reactive support

The greatest value of managed services comes from improving how collaboration environments perform over time. An effective provider should deliver more than service-level agreements and troubleshooting. It should analyze recurring problems, identify underlying causes, and recommend improvements that reduce future disruptions.

For example, repeated failures involving the same technology might warrant configuration changes, upgrades, or replacement planning. Providers can also help coordinate changes across internal IT groups, particularly when network modifications affect collaboration systems.

The ultimate measure of success is the employee experience. As Klosinski explains, “It’s not the ticket count, it’s the, you know, we’re helping your business work seamlessly.”

This proactive approach allows IT leaders to focus less on recurring operational issues and more on initiatives that support business growth.

Building a model that scales with the business

Collaboration requirements rarely remain static. As organizations expand or consolidate, adopt new platforms, and adjust staffing, managed services can provide the flexibility to accommodate these changes without continually rebuilding internal support capabilities.

AVI-SPL offers both remote-centric and on-site managed services for AV and UCC environments, allowing organizations to select an approach aligned with their operational requirements and network access policies. For CIOs evaluating providers, the priority should be a long-term partnership built around technical expertise, transparent performance reporting, shared operational standards, and continuous optimization.

As collaboration technology becomes more deeply embedded in business operations, the ability to manage it effectively will increasingly depend on having the right expertise, processes, and flexibility in place.

Learn more about how AVI-SPL’s Enterprise Managed Services can help your organization simplify AV and UCC management, improve collaboration experiences, and continuously optimize performance at scale.

For the past two years, the conversation around AI has focused on which industries it will disrupt. But AI’s rapid growth is also creating a major infrastructure challenge. As investment pours into data centres and supporting infrastructure, the AI boom is becoming central to delivering the next wave of the digital economy, reshaping construction in unexpected ways.

Earlier this year, NVIDIA CEO Jensen Huang called AI the “largest infrastructure buildout in human history.” For Australia, that presents a significant opportunity. Global tech companies are investing heavily in the infrastructure needed to support AI growth, and Australia is increasingly viewed as the place to do it because of our land availability, stable institutions and long-term renewable energy potential.

Commonwealth Bank has described Australia as the world’s third-largest destination for AI investment, behind the United States and China, with the country’s data centres pipeline estimated to reach A$150 billion by 2030.

The interesting part isn’t just the scale and opportunity AI presents. It’s that construction has reached its own inflection point, with the data and AI boom creating a rare opportunity to modernise how the industry plans, designs and delivers projects.

Construction is becoming more data-driven

Construction will always be grounded in the physical work of building.

But increasingly, data is shaping how projects are planned, coordinated and delivered.

Long before construction starts onsite, projects are developed in highly detailed digital environments. On the most advanced projects, buildings are now designed, coordinated and resolved virtually before anyone steps foot on site. Architectural, structural and services information is combined into a single living digital model where thousands of potential clashes can be identified and addressed up front, providing greater certainty of programme, cost and delivery.

This shift has made digital pre-construction one of the industry’s most valuable investment areas, reducing rework, improving productivity and minimising disruption for surrounding communities.

Unlike many other industries, construction generates large volumes of real-world data that general AI models can’t scrape from the internet. Laser scans and site cameras capture a detailed picture of how physical assets are designed and assembled, and companies that use that data effectively can build faster and with better certainty.

That data has historically been fragmented and underutilised.

AI is creating the opportunity to connect that information and turn it into better decisions, creating a significant competitive advantage.

The workforce shift no one expected

One of the more important opportunities created by AI is the chance to rethink how technical and trade-based careers evolve.

Construction and technology have typically followed separate talent pathways. Today, those worlds are increasingly converging as digital tools, data and advanced manufacturing become further embedded across the industry.

The workforce is beginning to evolve as modern construction becomes more digital. Tradespeople are increasingly working from digital models on tablets, assembling components prebuilt in factories and engaging with advanced manufacturing processes that would have been unfamiliar a decade ago.

At the same time, construction companies are recruiting machine learning engineers, AI product managers and advanced manufacturing specialists alongside traditional project teams.

The result is an industry with the potential to look very different from the stereotypes that still shape how it is often perceived.

AI still needs builders

Technology companies receive much of the attention when it comes to AI. But AI growth cannot exist within a vacuum.

Behind every model, agent and application sits an ecosystem of physical infrastructure that must be financed, designed, powered and built. As AI investment accelerates, the industries responsible for delivering that infrastructure will become increasingly important to the success of the digital economy.

There’s an irony in that.

The technology many people believe will transform the future relies on an industry still often viewed as traditional. For construction, that creates a rare opportunity: to use the data and AI boom not only to deliver the next wave of infrastructure, but to accelerate the industry’s own transformation.

AI does well in digital settings with clear rules. The physical world is messier and harder to predict. Ultimately, AI still relies on people to make decisions and safely build what it runs on. There’s something reassuring in that, and a little ironic. For all the talk about machines, it’s still people who set the pace.

DeepSeek has been dealt a huge blow in the US, even as it continues to innovate and keep pace with, or in some cases outperform, rival models from American labs.

The US Secretary of Homeland Security has issued an order banning all DeepSeek products and services from federal agencies. Government organizations must purge their IT systems of it within 60 days; those with “compelling justification” can make a case for exemption by written request to the Cybersecurity and Infrastructure Security Agency (CISA).

The Department of Homeland Security (DHS) cited multiple concerns about the Chinese-hosted model, including data sovereignty, lax Chinese laws, and security vulnerabilities.

Analyst Erik Avakian, technical counselor at Infotech, understands the concern. “When organizations use any type of hosted AI service, they’re potentially sending intellectual property, confidential business information, or other sensitive data into an environment where they may have very little visibility or control over what happens to it,” he said.

Government concerns

According to a document published in the US Federal Register, all technology developed by China-based Hangzhou DeepSeek Artificial Intelligence Co. will be banned from procurement by federal civilian agencies. This includes affiliated products and any third-party tools or software built using DeepSeek implementations.

Affected civilian executive branch agencies include the Departments of Treasury, State, and Justice; it does not apply, however, to private consumers, the general public, state/local governments, or Department of Defense (DoD) national security systems.

“DeepSeek’s models present multiple threat vectors,” the Order stated. To begin with, the company’s servers are located in the People’ s Republic of China (PRC), whose legal system grants “extensive authority” to government cyber and intelligence functions to access private infrastructure.

DeepSeek’s account generation process collects personally identifiable information (PII) and may also track and digitally fingerprint users, the Order asserted. Its prompt response process may result in the transfer of sensitive data to the PRC. This could give it “actionable intelligence” on US programs or individual users.

Further, the Order stated, DeepSeek models themselves may contain malware or other security vulnerabilities. DHS pointed to open-source security research published in February 2025 that uncovered many iOS mobile apps with multiple critical vulnerabilities.

Until the ban takes full effect, DHS advised agencies to mitigate risk by monitoring inbound and outbound data traffic, implementing enhanced logging and controls on data exfiltration, and detecting and flagging anomalous activity.

Agencies can submit requests for exceptions to the ban in writing, and must provide a “compelling justification,” such as its impact on mission-critical functions, or other considerations related to national interest such as national security reviews, investigations, or security agreements, as reasons why such exemption should be granted,.

Security issues are rampant

The ban comes just as DeepSeek is open-sourcing DeepInfer, its in-house interference framework. DeepInfer powers DeepSeek-V4.1-Flash with what the company said is high throughput and low latency. It features high-performance kernels, an optimized inference engine, structured outputs, and speculative decoding.

DeepSeek said its production token encoding and decoding components, which are aligned with processing logic behind the DeepSeek API, will bring the “same API behavior to community deployments.”

“Open models. Open kernels. Open inference,” DeepSeek posted to social media site X. “From private deployments to public clouds, we’ll keep working with the community to make large-scale inference more efficient.”

This isn’t its first roadblock from a cybersecurity perspective. Most notably, in January 2025, the company leaked more than one million log entries, including chat histories, API secrets, and other sensitive information, through an unsecured database.

DeepSeek reportedly addressed the exposure, but it still raised questions about the overall security practices surrounding the platform, particularly for organizations considering entrusting it with their own sensitive business information, Avakian pointed out.

That’s not to imply that every DeepSeek model is “inherently malicious,” he said, but there are still major questions around data privacy, the potential for information access under Chinese laws, the security of the broader AI ecosystem, and the susceptibility of DeepSeek’s models to jailbreaks and manipulation.

“Taken together, [these issues] raise legitimate concerns about whether organizations can adequately understand and manage the risks involved,” Avakian said.

Enterprises should seriously assess associated risk

Enterprises are increasingly exploring DeepSeek, although most publicly-documented adoption has been in China, Avakian noted. Some DeepSeek models have also been made available through platforms like Microsoft Azure and AWS.

“There’s genuine enterprise interest in the technology, as organizations look for opportunities to improve productivity, offer new capabilities, or reduce costs,” Avakian said. Useful applications include document summarization, internal research, and coding assistance.

DeepInfer also has the potential to reduce costs, but possible benefits should be weighed against associated security risks, he pointed out. The framework was just released, so organizations are going to need to validate the claimed efficiencies in their own environments rather than assuming savings will automatically translate into lower operating costs.

He encouraged organizations to start with relatively low-risk use cases, preferably using “public, synthetic, or appropriately sanitized information,” within a sandboxed environment they control themselves.

The important question to ask: ‘What is the business problem we’re trying to solve?’ From there, Avakian said, determine whether DeepSeek is the best way to solve it, validate security controls, and expand the use cases based on risk tolerance. Effective governance and risk assessment are key, he noted: understand what you’re bringing into your environment, validate its security, restrict its permissions, and make sure you can monitor what it’s doing.

“Approach this like any other software supply chain decision,” Avakian advised. “The fact that something is open source, inexpensive, or technically impressive doesn’t automatically make it appropriate for enterprise use.”

Impact on the broader AI ecosystem

The fact that the Order also addressed third-party products and services developed using DeepSeek implementations is important, Avakian said, because organizations may be using DeepSeek technologies indirectly, through vendors, software platforms, or other services, without necessarily knowing what’s operating underneath. “These are classic supply chain issues,” he observed.

It’s no longer enough to know which AI applications employees are using; enterprises must understand which models, frameworks, dependencies, and external services are embedded in products they’re purchasing and deploying, Avakian said. This means revisiting AI inventories, potentially updating procurement requirements, and expanding processes around third-party risk management, he advised.

“Ultimately, enterprises leveraging these tools need to understand not only the AI technologies they’re intentionally deploying, but also the technologies they’re inheriting through third parties,” he said.

Regulated industries have been dealing with the equivalent of sovereign data for a long time. And as the geopolitical situation makes it more broadly relevant, it doesn’t make it any easier to deal with.

From concerns about the reach of the US CLOUD Act and what telemetry or sensitive business data gen AI models might collect, to implications of the US government taking part ownership of a range of technology companies and the impact of EU regulations, the state of the world means more organizations need to think about where data is located.

The result is a flood of announcements from hyperscalers about new EU cloud regions, as well as pitches from regional and private cloud providers. The promised sovereignty may be anything from a local availability zone to more sophisticated approaches to ownership structure, compliance frameworks, where logs are stored, or where support staff work.

Actual data sovereignty involves legal, technical, contractual, governance, and infrastructure questions, but all too often, what’s on offer is what IDC research director Rahiel Nasir calls sovereignty washing. “When you look under the lid, all they’re really offering is a solution for data residency and data localization, keeping data in country,” he says. “That’s part of sovereignty, but not all of sovereignty.”

In his experience, in-country data centers are the first thing organizations ask for to support data localization and data residency, but they also require enhanced controls over identity and access management. That’s critical because data sovereignty is as much about who has access to the data and why, as where it lives.

“It’s really important the only people who touch my data in what is supposedly a sovereign cloud environment are those I give permission to,” he adds. “Because it’s my data, and I’m the only one who can give the authority to grant access to it. That also applies to support and admin staff.”

Companies want control through where they store data but also to restrict how users can interact with it that might expose that information outside an approved environment or jurisdiction, adds Acceldata SVP Ali Afsar.

Access and control

Don’t confuse the regulatory requirements of data residency with data sovereignty, Scality CMO Paul Speciale warns. “Residency is about geography,” he says. “Sovereignty is about control, jurisdiction, and who ultimately has the legal authority to access data.”

For CIOs, the question is no longer simply where the data is but who can access it, who controls the encryption keys, under which legal framework it operates, and can that control be proven. That includes being able to show who accesses data, from where, and for what purpose, which requires verifiable controls like jurisdiction-aware data placement, immutable backups, tightly governed admin access, and clear audit trails.

The UK government’s recent call for evidence on the impact of international data flows on sovereignty makes this explicit, referring to transfer by access where personal data remains in the UK but is accessed from overseas, including by a user located in another country. Access transfers include administrative access, backups, disaster recovery, support incidents and sometimes — depending on the specific regulations — processing the data.

GDPR doesn’t mean data has to stay in the EU but it does require data to be protected to EU standards wherever it goes. Data transfer obligations for other jurisdictions can be far more restrictive, so multinational organizations may need access controls limiting who can see what regulated data from different locations.

RBI-regulated payment-system data, for instance, must be stored in India, but can be processed outside India, Afsar explains, as long as it’s deleted from overseas systems within 24 hours. And with limited exceptions, personal data in Saudi Arabia must be processed within the Kingdom, and China treats overseas access to some categories of data stored in China as a cross-border data activity. Russian laws require similar strict data localization, and personal data of Russian citizens has to be stored, processed, and kept within the country’s borders at all times, which is why LinkedIn was blocked there.

Elsewhere, regulations are most restrictive for specific markets like healthcare, financial, and public services. Not everything needs to be transferred to a sovereign cloud, Nasir points out, as use of that data is subject to regulatory compliance, and the increasing need for AI training.

“This is the data that’s going to have catastrophic business impacts if leaked,” he says. “To give enhanced controls and security for that data, you’d consider wrapping sovereign controls around it. Everything else not subject to compliance nor sensitive data can be accessed by whomever you want and from wherever.”

Sovereignty and safety

Organizations typically know data sovereignty rules well but still experience incidents like data breaches, government data access requests, and unauthorized cross-border transfers caused by, for example, traveling employees or excessive admin access, according to Kiteworks’ 2026 Data Security and Compliance Risk report.

“Location only tells a regulator where the data rests, not who can still reach it,” says Kiteworks’ EMEA GM Dario Perfettibile. “Every real exposure we’ve seen so far happens through access not geography.”

Thinking data needs to stay in region because then it’s safe gives a false sense of security if practiced as a substitute for proper data management. Access, data labelling, and retention polices take more work but are what can prevent incidents. “If you have a data policy in place, it can force a reapproval of the access,” he adds. “It can restrict access to you only, or it can block you out immediately.”

Location isn’t everything

Using a service closer to home doesn’t automatically make any of that easier. “Many organizations still assume sovereignty can be achieved by moving data from one cloud provider to another, or by replacing one hyperscaler with another,” says Keepit Group CISO Kim Larsen. “That misses the bigger picture.” CIOs should treat sovereignty as an operational capability not a compliance exercise, he adds, since sovereignty isn’t about scale or ownership of infrastructure for its own sake, but maintaining control.

“If an organization doesn’t know who can access its data, which legal frameworks govern that access, or whether it can recover its data independently, simply changing where the servers are located does little to reduce risk,” he says.

Just because something is based in your region doesn’t automatically make it a good choice. The EC, for instance, has set up accounts on W Social, an independent social media platform based on the AT protocol designed for data mobility and user migration, which Bluesky uses but is hosted in Europe.

The Commission’s own open source strategy suggests open source promises more control over digital infrastructure by avoiding vendor lock in, and a recent policy brief positions digital sovereignty as a form of digital independence. But having servers in the EU isn’t the same as freedom of digital movement since unlike Eurosky, based in Europe and using the same protocol, W limits interoperability so it can require identity verification, and has closed its source code after security researchers pointed out vulnerabilities.

Most organizations will continue to use both global and local providers, though they might prioritize what Nasir called glocal options built on partnerships between the two. “Organizations want to use a mixture of local and global providers,” he says. “They want all the innovation capabilities that the Microsofts, AWSs, and Googles of the world give us, that they consider local providers can’t match.”

IDC figures suggest only 4% of European organizations plan to switch to purely local options. “We can’t live without US technology today and there’s no appetite to do that,” says Nasir. Availability matters and local providers may not have the same level of infrastructure as global options. They can offer low hanging fruit like video conferencing solutions, but not more complicated directory and database services, or productivity software that’s intertwined with those US-provided services.

Choices about where to store data, and the technical controls to enforce that, have their own operational implications.

“Redundancy and resiliency are key tenets of sovereignty,” he adds. “You want to make sure your operations are always available. In order to enable that, you want to spread the risk.”

Stripe recently acquired AI routing platform OpenRouter for a reported $7.5 billion, making it one of the largest venture-backed AI startup exits and fintech M&A deals of the year. It’s a blockbuster deal that instantly became one of the most intriguing tech developments to occur this year. Not just because of the dollars involved or the reasons your tech newsfeed or social media timeline insists.

The broader perception seems to be that this is just a standard software developer tooling narrative. Financial analysts and tech enthusiasts would have you believe the main takeaways here are the massive acquisition price and Stripe’s shiny new resource to lower AI token costs. To many, OpenRouter is simply a tech-savvy developer utility that routes basic prompts to inexpensive AI models while setting aside pricey frontier models for complicated reasoning tasks.  Others view it as a way to prevent vendor lock-in from cloud ecosystem giants such as AWS Bedrock. But these perspectives are missing the big picture.

From my perspective, it’s clear that this acquisition sends a definitive signal about where the future of digital payments is heading – not just the future of artificial intelligence software. While it won’t happen overnight, agent-to-agent commerce is clearly opening the floodgates to a paradigm where autonomous software agents will serve as fully empowered economic participants that can conduct business on behalf of both individuals and enterprises.

In my line of work, that means we aren’t far away from the time when, for example, an automated car insurance agent can eventually meet, negotiate with, evaluate risk for and sell coverage to another company’s autonomous agent – with the entire exchange happening seamlessly behind the scenes.

From agents that assist to agents that transact

Consider how AI agents operate nowadays versus how they will perform in the near future. Today, agents primarily pass context back and forth across software interfaces. Case in point: A consumer uses a car shopping agent to search vehicle inventory for a new Toyota RAV4; after the consumer chooses the exact vehicle they want, the agent passes the pertinent details – make, model, buyer location and other criteria – to an insurance-quoting agent. In this model, data flows fluidly. Yet the software systems themselves complete no actual financial transaction. The agents aid the user, but human intervention via a checkout process, credit card entry or manual sign-off is needed to finalize the vehicle purchase.

Meta’s recent move to open access for developers to build Muse connectors is proof that this transition from passive assistance to direct execution is already underway. Personal agents like Muse, Instinct, OpenClaw and Dots can now leverage simple API connectors to handle browsing and user context in the background. At the same time, e-commerce giants like Shopify are giving agents direct access to checkout, with OpenRouter serving as the critical routing engine sitting right in the middle. This demonstrates that the real secret sauce is in the transaction layer, not just model intelligence.

These developments signify a major leap from agent assistance to direct execution. In the agent-to-agent economy just over the horizon, software agents will actively complete end-to-end transactions without an information handoff hard stop. Instead of multiple agents involved, a single AI agent will be able to buy specialized micro-services from another agent, retrieve proprietary data from secure databases, negotiate contract details using preset operational guidelines and fulfill sophisticated tasks on demand.

Before panic ensues, let me make it clear that this shift doesn’t mean real people, or their jobs, will completely disappear from the loop. This entire ecosystem still requires oversight from human beings, who will be needed to set important goals, define ethical boundaries and enforce regulatory safeguards. The fundamental change underway is that the machine-to-machine layer will be increasingly responsible for the mechanics of commerce, and the minutiae humans find tedious.

Agent-to-agent commerce isn’t just a niche tool or a standalone use case. It’s a new layer of interaction, one that could operate just beneath the surface of many existing business processes and revamp how enterprises negotiate, exchange services and produce revenue.

The infrastructure problem

None of this will be easy. Reconfiguring an economy centered around autonomous software agents presents colossal financial and technical challenges that conventional banking systems weren’t designed to navigate. Comprehending the difficulty ahead requires looking more closely at the basic mechanics of machine-to-machine interactions. For starters, software agents interacting in real time lack built-in trust capabilities. They need instant, verifiable financial settlement to validate that tasks have been finished before they can permit access to data or services. What’s more, the dollar amounts involved in these micro-payments are commonly too tiny to validate invoicing methods or standard payment processing workflows.

And therein lies the rub, and the infrastructure gap laid bare: Existing payment systems simply weren’t built for machine-speed, high-volume micro-transactions. Credit card networks, wire transfers and other traditional payment pathways depend on static batch processing, multi-day settlement windows and high fixed transaction fees. For instance, when an AI agent must make a $0.002 micro-payment to another agent to process a 1,000-token prompt, access a database or perform a targeted computation, waiting two business days for an ACH transfer or forking over a $0.30 credit card processing fee makes the whole interaction economically infeasible.

Of course, this infrastructure problem isn’t happening in a vacuum. Significant tech and digital finance players are working furtively to create machine-native transaction systems. Exhibit A: Fintech innovator Circle is crafting stablecoin payment systems that enable AI agents to send and receive funds in real time without human intervention needed. Exhibit B: Joint initiatives are being developed, like the agentic commerce protocols by Stripe and OpenAI. These and other examples prove that industry movers and shakers acknowledge that empowering machines to safely and instantly transact money represents the next crucial frontier for digital infrastructure.

Viewing Stripe’s acquisition of OpenRouter through this lens totally reframes the deal as not just a straightforward software purchase, but a major step in a multi-pronged infrastructure strategy. Think of OpenRouter as a central dispatch hub that oversees and directs scores of AI agent workloads throughout many different systems. It links developers to legions of AI models across numerous providers. And it carefully assesses every incoming request to pick the best model based on cost, speed and complexity of the task. Stripe, on the other hand, serves as the international financial backbone in this dynamic, providing stablecoin rails and the micro-payment wherewithal necessary for real-time monetary settlement.

Like two gigantic pipes that thread snugly together, Stripe and OpenRouter become the missing plumbing for agent-to-agent commerce. OpenRouter functions as the master control valve in this metaphor, directing where tasks go; Stripe then supplies the high-pressure financial flow that instantly settles the value exchanged.

It’s this synergistic dynamism that helps the estimated $7.5 billion price paid make sense. This isn’t merely an existing software business with a dependable revenue stream Stripe is absorbing; the new parent company is securing a controlling position at the precise historical inflection point where AI workloads pivot to autonomous commerce. Stripe is now strategically placed at the intersection of digital payment settlement and model orchestration. Consequently, as AI agents emerge as primary economic drivers, every token request and micro-transaction is routed through its combined plumbing.

What this could mean for insurance

All of which brings us back to my domain: the insurance industry, a space where transactions already include multiple parties, ultra-complex data structures and meticulous choices made across narrow organizational boundaries. Currently, manual underwriting, multi-step risk evaluations, document verification and legacy payment clearance are required to bind an insurance policy. But envision this paradigm in an agent-to-agent economy. Revisiting that RAV4 buyer, imagine their personal car-shopping AI agent verifying one available for purchase at a nearby dealer and then instantly contacting an insurance company’s autonomous agent. The latter examines the vehicle specs and driver history, performs underwriting computations, calculates a precise risk premium, validates data credentials, collects the premium using instant stablecoin or micro-payment rails and enacts the coverage mid-transaction – all with zero human assistance.

This is more than a hypothetical scenario. It’s an emerging reality, one that requires immediate attention from carriers and enterprise leaders. It means insurers must now develop algorithmic underwriting logic that a machine can apply and defend, construct verifiable data exchange systems that guarantee authenticity and data privacy during AI negotiations, and implement financial methods that can manage instant premium settlement 24/7. They must also begin planning for how their product gets represented, discovered, evaluated and transacted by someone else’s agent. Because the foundational infrastructure for this autonomous economy is no longer theoretical. Companies are already building the routing, identity, payment and commerce layers that could allow agents to become economic actors.

The ultimate question is not if or exactly when this machine-driven sea change will arrive. It’s what enterprises can do right now to ready their products and platforms to surf these waters successfully, catch the rising wave and ride it out with perfect poise.

The AI industry says enterprises can’t absorb AI. Its answer is to absorb it for them.

That’s not the same thing.

The engineer across the table knew the model cold. Not vendor-deck cold. He could tell me where it would break, what it would cost, and which benchmark numbers to ignore. One of the better technical minds who has ever sat in that chair.

Forty minutes in, I asked which of our problems he would point to first. He named a minor workflow one of my directors had mentioned that morning.

The easy read is that he didn’t know my business. True, and not very interesting. Here is the real issue: the people who sent him are making the exact mistake they are being paid to fix.

They’re not wrong about the problem

The frontier labs aren’t confused, and I want to be fair before I disagree.

MIT found that about 95% of enterprise AI pilots produced no measurable profit impact, despite tens of billions spent. Winning the benchmark wasn’t winning the deal. And a customer who never gets to production never renews.

So, they did the logical thing. They sent engineers:

• OpenAI: Launched the OpenAI Deployment Company in May with $4 billion of initial investment from 19 partners, starting with 150 forward-deployed engineers (FDEs) and deployment specialists.
• Anthropic: Partnered its engineering capability into an enterprise AI services venture with Blackstone, Hellman & Friedman, and Goldman Sachs, reportedly valued above $1.5 billion.
• Microsoft: Launched Frontier Company in July with $2.5 billion and approximately 6,000 embedded engineers and industry experts.
• AWS: Committed $1 billion to a forward-deployed engineering unit embedding engineers with customers.
• Market surge: FDE job postings grew roughly 800% in nine months, with principal-level compensation past $1.2 million.

That is a serious response. But it answers the wrong question.

All that capital is aimed at getting customers into production and keeping them there. We have been reading it as the playbook for getting AI into the wider economy. Those are two entirely different problems, and only one of them currently has anyone working on it.

The same mistake, twice

One structural fact tells you what this really is: the forward-deployed function sits on the commercial side of these companies, not in research. OpenAI capitalized its deployment arm as a separate business with outside investors.

That’s not an accusation; the org chart is simply more honest than the positioning. What that structure measures is pilots converted to production within the quarter. Against that scoreboard, the model isn’t shallow at all. It’s optimal.

Now look at the same failure from my side of the table.

A team builds an intelligent demand-forecasting agent. Accuracy improves by double digits. The engineering readout is genuinely good. And absolutely nothing happens to cash.

Accuracy was never the point. It only matters if it changes an operational decision: how much safety stock you hold, what commitments you make to suppliers, or what you schedule to build. If none of that changes, you haven’t freed a single dollar. Worse, if the model was tuned for fill rate, because that’s what the commercial team asked for — inventory climbs, working capital gets suffocated, and you are worse off than before. The project wins. The company loses.

Anyone who runs S&OP knows this reality: inventory is cash sitting on a shelf. Every forecast is a claim on that cash. Every service-level promise decides how much liquidity you lock away. The value is never in optimizing a single, isolated number; it is in managing the trade-offs between them.

Outside engineers cannot own those trade-offs. Look at what their engagements are sold on: a deflected ticket, an automated records request, a closed underwriting case. Local, siloed inside one function, and measurable in weeks. Those are the only outcomes an outsider can deliver before rolling off to the next logo.

The results a CFO cares about cross three functions and only show up in cash two quarters later.

Delivery vs. Diffusion

This gets celebrated as AI reaching the real economy. In practice, it does close to the opposite:

• The learning runs one way. Anthropic’s job listings are the most explicit about this, asking engineers to codify repeatable deployment patterns and contribute those insights back to product and engineering. That is honest, and it is the point. Analysts have flagged the same concern from the buyer’s side: that deep reliance on frontier labs hands those labs insight into the very industries they may later enter.
• Capability is replaced instead of built. The engagement leaves behind a functioning workflow, but zero internal muscle to solve the next bottleneck. Constellation Research warned this year that FDEs are often used as a crutch for product immaturity, and as part of a broader vendor dependency play. Without Palantir’s feedback loop — where every deployment continuously enriches an integrated customer ontology — this is simply an expensive systems consultancy with a frontier brand name.
• The mid-market waits. Dice reported in August that the shortage of engineers is now directly holding back enterprise AI returns. The scarcest engineering talent sits inside Fortune 50 anchor accounts, while the companies that make up most of the economy queue behind them.

In engagements structured this way, the lab learns your business. You learn their product.

So here is the test I now apply to any engagement, including the ones I sign: when it ends, are we better at finding the next problem ourselves?

If the capability leaves the building when the outside engineer packs up their laptop, that was delivery, not diffusion. Delivery is a fine business, but it doesn’t compound. Diffusion is nothing but compounding.

What we did instead

We eventually fixed our inventory problem.

Not by pointing an off-the-shelf frontier model at raw ERP tables, but by redesigning our entire planning architecture from the ground up. We staffed the initiative with operators who had spent years inside our supply chain and knew where the real constraints actually sat — the lead time, the allocation rule, the supplier commitment that no outsider would have known to model and then applied AI directly to those decision points.

The result: Inventory came down and stayed down, turns improved, and the part that matters for this argument, nobody had to be flown in to do it again the following quarter.

There is no version of that outcome a single outside engineer could have designed, no matter how brilliant the model in his hands.

The bigger stake

Scale this up and the macro picture becomes clear. A country’s AI advantage will not come from a few hundred flagship deployments. It will come from millions of operators applying AI to operational decisions they were previously making badly.

Which brings me back to the man across the table.

We are spending millions sending AI experts to learn business. We should be sending business operators to learn AI.

• Business knowledge is slow. It took most of my career to instinctively know which trade-offs to make around working capital and inventory.
• AI execution is fast. An experienced operational team can master agentic tooling, evals, and model orchestration in a matter of months.

We are running the harder transfer at immense cost and mistaking it for progress.

I don’t need more forward-deployed engineers. I need backward-deployed domain experts.

Nobody is selling those. Which means the only people who can build them are already on your payroll.

Per-token pricing has become the default commercial language of generative AI, and practically every budget conversation now happens in those terms (tokens consumed, tokens generated, cost per million tokens, etc., etc.).

The problem is that this model treats every AI interaction as a discrete transaction, separate from the infrastructure required to process it. Yes, tokens are a billing unit, but the work they represent also has a physical footprint in memory and storage, and as agentic AI drives context windows into the hundreds of thousands of tokens, that footprint has outgrown the GPU.

So why does this matter? Well, according to recent analysis published in Harvard Business Review (HBR), corporate AI is about to experience a massive “cost shock”, as vendors move away from unmetered or bundled AI features and begin passing on the cost of the infrastructure required to deliver them.

As HBR points out, organizations are currently being paid, in effect, to adopt AI, with major software vendors quietly absorbing the steep costs of GPUs, inference and tokens on their customers’ behalf. That subsidy is not built to last.

Take memory and storage costs, for example. DRAM prices have risen sharply as manufacturers focus on High Bandwidth Memory (HBM) production for AI accelerators, tightening supply for conventional server memory. Meanwhile, demand for AI is exerting sustained pressure on the availability and pricing of high-capacity SSDs.

This is totally unsustainable, and there was always going to come a point at which vendors could no longer continue covering these costs themselves and would need to pass more of them on to customers. That point is now rapidly approaching.

The scale of the shift is hard to overstate. Google was processing 9.7 trillion tokens a month in 2024; by mid-2025 that had risen to 480 trillion, and at this year’s I/O, Sundar Pichai put the figure at 3.2 quadrillion tokens a month. Goldman Sachs projects industry-wide token consumption will multiply another 24 times by 2030. Whatever an organization budgeted for inference last year, the volume assumption behind it is already out of date, and a growing share of that volume is input, not output: agentic workflows that plan, call tools and re-read context on every step are pushing input tokens, not generated ones, to the top of the cost stack.

Your AI budget has a serious NAND problem

The clearest sign of where this is heading came at CES 2026, when Nvidia unveiled what it first called its Inference Context Memory Storage Platform, built around its coming BlueField-4 DPU, and has since formalized as the NVIDIA CMX context memory storage platform. CMX standardizes the offload of the model’s working memory, the KV cache, onto NVMe SSDs, extending GPU memory into a shared, pod-level context tier and making it persistent across inference runs. Nvidia says the approach can deliver up to five times greater power efficiency and up to five times higher tokens-per-second throughput, and it already has more than a dozen storage partners lined up to support it, including Dell, IBM, NetApp, VAST and WEKA.

That is a significant validation point: several storage vendors were already extending GPU memory onto flash before ICMSP existed. Nvidia standardizing the approach across its ecosystem confirms this is now an architectural direction for the industry, not a niche workaround. But it also means the economics of inference are no longer determined solely by GPU pricing; they are now directly exposed to storage media pricing.

That exposure matters because flash pricing is anything but stable. According to VDURA’s own Flash Volatility Index, 30TB TLC SSD pricing rose 472% between Q2 2025 and Q1 2026, and the price multiple between high-capacity QLC flash and hard drives widened from roughly 4.9x to more than 22x over the same period, while HDD pricing moved about 35%. If cost-per-token now depends on flash-resident context tiers, then token costs inherit flash volatility, and organizations running all-flash inference infrastructure are carrying unhedged exposure to the NAND market on every query.

Instead, flash should be reserved for hot context data where its performance genuinely affects inference. This is because not all types of data generated or retained by AI workloads have the same latency requirements. For instance, data retained for future use, such as checkpoints, does not necessarily need flash performance. The hyperscalers worked this out long before the current price cycle: Google, Meta and Microsoft do not run all-flash storage; they run mixed fleets, with just enough flash to saturate compute and intelligent tiering draining everything else to high-density disk. A mixed-fleet architecture separates performance capacity from bulk capacity, giving organizations a structural hedge against volatile flash prices.

Tread carefully

If they aren’t doing so already, infrastructure leaders should model token-cost sensitivity to media pricing before committing capital. Do not forget that agentic workloads can disproportionately increase token consumption on the input and context sides. As a result, cost models must account for how much context requires high-performance flash capacity and how much can be stored on lower-cost tiers. Teams should regularly test how projected AI costs change as CMX-style architectures push more context onto NVMe, and as SSD pricing rises further while context requirements grow.

That all being said, HBR’s concluding analysis is extremely apposite: as long as executives are being asked to redesign their organizations around a technology with unpredictable pricing, they should tread carefully.

Over 20-plus years in the enterprise trenches, I’ve watched multi-million-dollar technology transformations stumble over the same blind spot. We spend months debating cloud architectures, database benchmarks and vendor feature matrices. But the tech itself? That’s rarely where the real bleeding happens. The real pain arrives later, in the messy organizational wake:

• Who actually controls this system?
• What critical business function did we just accidentally outsource?
• How painful will it be to untangle ourselves when the vendor changes their licensing model?

Agentic AI is about to make those questions considerably more interesting. We have spent years worrying about application and vendor lock-in. The next form of lock-in may sit several layers deeper.

For most of my career, and arguably most of the history of computing, machines have waited for us. We gave the instructions, approved the transactions, made the decisions. Agentic AI is beginning to change that relationship. An agent can interpret an objective, break it into tasks, find the services it needs and execute those tasks with limited human intervention. The moment an agent can do more than recommend a purchase and can actually make one, it stops being a technology tool. It becomes an economic participant.

Few technology leaders are asking the question that follows — and by the time more of them do, the answer may already be decided: What financial infrastructure will these machines operate on, and who will control it?

The International Monetary Fund’s 2026 note on agentic AI and payments describes a shift from human-initiated transactions toward agent-mediated ones — from a “click-to-pay” model to a “decide-to-pay” model — with implications for authorization, settlement, liquidity, compliance and resilience. The IMF also highlights a fundamental tension between probabilistic AI behaviour and the deterministic requirements of payment infrastructure.

That caution is worth sitting with. If the institutions closest to global payments infrastructure are already examining how to preserve authorization, control and settlement finality as AI becomes more autonomous, that is itself a signal about how much is at stake in who builds and governs the rails.

Picture what this looks like inside an enterprise. A procurement agent instructed to reduce the cost of a service identifies several providers, checks their credentials and reputation, negotiates terms and initiates payment — with a human approving only the exceptions. A logistics agent dynamically buys transport capacity as demand shifts. A software agent discovers an API, uses it for a specific task and pays for the consumption without anyone visiting a payment page. None of this requires exotic technology. It requires identity, authority, trust, reputation, marketplaces, wallets and payment rails built to work with autonomous participants rather than humans clicking “confirm.”

In other words, the agentic economy will need its own economic infrastructure.

Money is becoming infrastructure

The petrodollar is the clearest historical parallel, and it’s worth being precise about the mechanism rather than just the label. In the early 1970s, as the dollar came off the gold standard, the US reached an understanding with Saudi Arabia, and by extension OPEC, that oil would continue to be priced and settled in dollars. Oil-exporting nations then recycled much of that revenue back into US Treasury securities and dollar-denominated assets. Nobody decided that oil needed a new currency invented for it. The currency that was already dominant simply became the substrate the world’s most strategic commodity traded on — and countries with no direct relationship to the US found their energy trade, and much of their national accounting, running through a currency they didn’t issue and couldn’t control.

The analogy to AI is imperfect but suggestive. If compute and intelligence become this century’s strategic commodities the way oil was for the last one, could the infrastructure through which autonomous machines exchange value become dominant in the same quiet way?

Pieces of this are already forming. The US has built a regulatory framework around dollar-backed payment stablecoins, and the US Treasury has explicitly linked implementation of the GENIUS Act to reinforcing the dollar’s role in digital finance. China has the digital yuan, Europe is progressing a digital euro and BRICS countries are exploring local-currency settlement and alternative rails. None of this makes an “AI dollar” inevitable, and these initiatives aren’t all competing for the same role. But they show money becoming part of the digital infrastructure itself, not just a thing that moves across it.

The Bank for International Settlements’ 2026 Annual Economic Report found that more than 99% of fiat-backed stablecoin supply remains dollar-pegged, even though stablecoin use in the real economy is still small next to traditional payment systems. That gap matters: The infrastructure is developing quickly, but the machine economy hasn’t been built on top of it yet. We’re unusually early — discussing the architecture before it becomes hard to change.

Two paths, and probably a third

There are broadly two ways this could evolve. The first extends today’s financial architecture into the machine economy: Agents get digital identities, wallets and delegated authority, but ultimately still operate through the banks, payment networks, financial institutions and technology platforms we already use, with an AI layer added on top. There’s nothing wrong with this. We already have institutions, regulation, compliance and dispute-resolution mechanisms built for handling money — reinventing all of that simply because the customer is now an AI agent would be wasteful.

The second, more interesting possibility is a more open economic layer built specifically for autonomous agents: Open marketplaces where agents discover services across providers, establish trust through portable identity and reputation, and transact using interoperable payment mechanisms. Trust doesn’t disappear in this model; it gets established through verifiable identity, reputation and programmable rules instead of a single central intermediary. I’m not suggesting everything should be decentralized, or that blockchain automatically solves this — financial transactions carry requirements around compliance, legal finality, fraud prevention and accountability that can’t simply be wished onto a distributed ledger.

The World Economic Forum’s work on AI agents and payments makes a similar point from the financial-services perspective: Traditional identity controls may no longer be sufficient when software is acting between the customer and the transaction. Understanding an agent’s intent, authority and context becomes part of establishing trust.

The fork that actually matters is this: Do we want autonomous agents to inherit the financial dependencies we’ve spent decades trying to untangle, or is this a chance to design something more open from the start?

It matters for enterprises specifically because we’ve lived this before. We’ve spent years reducing application and infrastructure vendor lock-in. Agentic AI could create a far broader dependency — not lock-in to a single application, but to an entire technology and economic ecosystem at once. This is what I’d call AI supply-chain sovereignty: Not owning every component of the stack but understanding the dependencies well enough to retain strategic choice. If your models, cloud infrastructure, agents, identity layer, marketplace and payment rails all sit inside one ecosystem, you may have built a sophisticated version of lock-in without ever naming it that.

In practice, the resolution is probably a hybrid rather than a clean win for either model: Regulated rails carrying legal finality, dispute resolution and compliance, with open protocols layered on top for discovery, portable identity and reputation. Picture agents shopping across an open marketplace for services, but settling and resolving disputes through accountable, licensed intermediaries’ underneath. That blend is the more likely third path — and probably the one enterprises should be designing toward now, rather than betting everything on either pure extreme.

What this means in practice

A few things follow from this for technology leaders. Make model and infrastructure portability an architectural principle, not an afterthought. Understand where private infrastructure and open-weight models can sit alongside commercial frontier models, and design agent architectures capable of talking to multiple services rather than being permanently wired to one marketplace.

This is increasingly consistent with how McKinsey describes the infrastructure shift required for agentic AI: Modularity, composability, decoupling and vendor flexibility become important as agents interact across systems, tools and transactional environments. For technology leaders, that is not simply an infrastructure concern. It is an argument for preserving optionality in the broader AI stack.

Rethink what an agent wallet actually is. The important concept isn’t cryptocurrency — it’s delegated financial authority. An enterprise agent needs something like identity and access management for money: Who is this agent, what is it authorized to purchase, how much can it spend, which counterparties can it use and when does a transaction escalate to a human?

Recognize that payments are no longer purely finance’s problem. For years, technology has treated payments as something that sits on the other side of the org chart: Finance decides how money moves, and technology integrates whatever’s required. Once a machine can decide and execute a transaction in the same motion, the payment mechanism becomes part of the technology architecture itself.

Finally, add the economic layer to your AI architecture diagram. We already map data, applications, models, cloud and cybersecurity. It’s time to also map where identity, reputation, marketplaces, payment rails and currencies sit in that chain — and what happens to the business if any one of them becomes unavailable. That’s the practical meaning of sovereignty here: Not owning everything but making sure no single component can switch off the intelligence the business depends on.

The choices we make now

The larger question is still open. We could reproduce today’s financial architecture for machines, with agents operating inside increasingly sophisticated ecosystems controlled by governments, financial institutions and platforms. Or we could deliberately build more open, interoperable economic infrastructure where agents discover services and establish trust across organizational and technological boundaries. The real outcome will likely be messier than either — regulation and trusted institutions aren’t going away, even as open protocols increasingly provide the connective tissue between autonomous participants. What matters is that we still have some influence over that balance.

The petrodollar wasn’t created because someone decided oil needed a new currency; it emerged from the interaction of a strategic commodity, global trade, financial markets and geopolitics. The agentic economy may develop the same way, with compute and intelligence generating the economic value and new financial infrastructure determining how that value moves.

For technology leaders, the question may end up being bigger than who builds the smartest AI. It may be who controls the economic infrastructure through which autonomous intelligence participates in the economy — and whether we make that infrastructure open enough that no single company, platform or jurisdiction becomes indispensable.

Every CIO in a regulated industry has a system the business depends on, that no one fully understands and that must be replaced. In banking, it is often a core platform that predates the internet. In my own career, that system took two distinct forms: first, a VAX minicomputer running a parking operator’s entire business across nearly 200 New York City locations; later, the on-premises core of a global financial technology company supporting payments, banking and card issuing. Both had to move to the cloud without interrupting the customers that depended on them. The technology differed each time. The reasons the migrations succeeded did not.

Why regulated migrations fail differently

Migration failures come in two kinds. The loud kind takes a website down, engineers fix it and customers move on. The quiet kind — a pricing discrepancy, a missed filing, a mismatched reconciliation or a lost audit trail — may not surface for weeks. Its cost is measured in regulatory findings and lost client contracts rather than in hours of downtime.

Regulated migrations face both. The loud kind can be ruinous, as TSB learned in 2018 when a big-bang platform migration locked customers out for weeks and, after regulators found insufficient contingency planning and untested cutover configurations, drew a £48.65 million fine. The loud kind is at least visible. The quiet kind breaks the business without anyone noticing.

This changes what “without breaking the business” means. It does not mean zero outages. It means that at every point during the migration, the organization can prove that the new system produces the same outcomes as the old one for the cases that matter, or a difference the business has approved, and has a tested way to recover if it does not. That standard is stricter than most migration plans are written to, and it determines whether the program survives its first serious incident. Meeting it comes down to five operational disciplines.

Map the dependencies you were not told about

The formal dependency map in the architecture repository is always incomplete. The dependencies that break a migration are the informal ones, which grew up around the legacy system without ever being documented. A team in another department builds a nightly job that reads a file the legacy system happens to produce at 2 a.m., without the system’s owners knowing. An analyst in finance writes a spreadsheet macro that queries a database table directly, bypassing every official interface. A report goes to a regulator every quarter, generated automatically for so long that no one remembers how to produce it by hand.

Finding these dependencies takes careful observation. Configure the legacy environment to record every connection, every file it produces or consumes, and every scheduled job. Collect this data over a full business cycle, including month-end and quarter-end processing, before relying on the dependency map. This step is often dismissed as unnecessary, but it provides the evidence needed to plan the migration sequence. A workload should move only when its dependencies have moved, remain reliably accessible, or are confirmed to be no longer required.

The map must also account for people. Behind every critical legacy workload is a group of engineers or business analysts who know why an undocumented parameter exists or why a batch job runs only on the third Thursday of the month. Identify them at the outset, protect their time and capture what they know in reproducible test suites before they retire or move on.

A thorough map also reveals what to prune. Over decades, legacy estates accumulate dormant batch jobs, deprecated extracts and orphaned interfaces. Retiring these before cutover shrinks the migration scope and removes testing cycles spent on workflows that no longer serve anyone. Never decommission an idle process based solely on the monitoring window; verify with the responsible business owner and check historical run schedules for low-frequency jobs first.

Sequence cutovers around the business calendar, not the project plan

A project plan wants cutovers on a steady cadence. A regulated business has dates on which nothing may change, such as settlement days, filing deadlines, rate resets and month-end close. Build the migration calendar from the business calendar outward, placing cutover windows where you can detect and recover from failure before the next immovable date.

The discipline that makes this work is proving it at every stage before scaling to the next. Start with one workload, such as monthly billing or a single settlement feed, in the incremental style the Azure Architecture Center calls the strangler fig pattern. Run it on the new system in shadow mode alongside the legacy system for a full business cycle. Keep the legacy system as the system of record during this period. Capture the new system’s intended actions, such as payments and client messages, without sending them. Next, compare both systems’ outputs for each transaction and resolve any differences. Test external integrations, peak-load performance and recovery procedures separately, since shadow running never exercises the actions it suppresses.

Only then make the new system authoritative for that workload. Retain a tested recovery path through an agreed stabilization period, with a method for handling transactions recorded after cutover, and move on to the next workload. Where the architecture does not allow workloads to move independently, apply the same discipline to a coordinated group of them, with rehearsed cutovers and explicit acceptance criteria in place of the workload-by-workload sequence. Where even shadow running is not feasible, establish readiness through representative transaction replay, end-to-end testing and full cutover rehearsals, against the same acceptance criteria and with the same tested recovery path.

This is slower than a big-bang migration by a factor that sometimes makes executives uncomfortable. It is also the approach that produces evidence, rather than hope, before each increase in production exposure.

Keep the old system running until the new one has earned trust

The strongest temptation in any migration is to switch off the legacy system early, because running two systems costs money and because the old one is what the program exists to eliminate. Resist it. In a regulated environment, the legacy system is the reference implementation. It is the baseline against which every output of the new system is evaluated. Explain and resolve every material difference between the two systems, never mistaking agreement for correctness; then check the results against the business rules and the authoritative records.

When we moved a parking operator’s business from a VAX minicomputer to a cloud platform, the VAX ran alongside the cloud platform for months, and we reconciled every transaction between them until the new system earned the business’s trust. The parallel run let us catch and resolve issues before they reached customers. The cost of the parallel run was visible and budgeted. Had we skipped it, our customers would have discovered the errors, such as mischarged accounts and failed payments, before we did.

Rehearsal matters at every scale. When Commonwealth Bank of Australia moved its core banking platform to the cloud in 2025, it conducted five full dress rehearsals of the cutover, having originally planned for three. During the actual cutover, it kept customers online throughout the three hours the core was down.

Reconciling parallel outputs exposes discrepancies, and not every discrepancy is a defect in the new platform. Shadow running frequently uncovers latent bugs in the legacy system, such as rounding oddities, edge-case truncations or mishandled time zone changes, that went unnoticed because no independent system had ever checked the arithmetic. Classify each discrepancy as a new-system defect, an undocumented legacy defect or a deliberate change in method. New-system defects must be fixed. Legacy defects need an impact assessment, remediation and, where required, disclosure. Deliberate changes require approval from the business and, where applicable, the regulator. Working through that classification transparently earns the consensus needed to retire the legacy host.

Reconciling transactions does not prove that the migrated data is complete. Before retiring the legacy system, validate balances, record counts and relationships between records, and confirm that historical records and audit evidence remain retrievable, whether in the new system or in a validated archive.

Build the controls into the platform, not around it

Regulated migrations often treat security, resiliency and auditability as a compliance workstream that runs beside the technical one. That is how gaps appear at cutover. Build the required controls into the target platform before the first workload arrives: automated data residency enforcement, encryption with customer-managed keys where the data classification calls for them, continuous monitoring, immutable audit trails and granular identity and access management. Every migrated system then inherits these controls from the start, and each workload’s use of them is verified. Compliance is not a post-deployment audit; it is a primary design constraint that dictates the perimeter of the technical solution.

This matters doubly when the platform is intended to host AI. Models and agents will act on the data the migrated systems hold. Audit trails must be immutable and retrievable for years, and decisions that affect customers will need an explainability that many AI systems do not provide by default. If data lineage, access controls and the audit trail are not in place at migration, retrofitting them afterward is costly and often incomplete. Evidence that was never captured cannot be reconstructed, and the AI program will inherit a foundation that cannot withstand regulatory scrutiny.

Govern with standards the business can read

Enterprise architecture frameworks, such as TOGAF, provide the architecture team with a process; industry reference models, such as BIAN, give the business a vocabulary. The Banking Industry Architecture Network (BIAN) publishes a standard model that partitions a bank’s work into named business capabilities, such as payment execution or customer agreement, each with a standardized interface. When migrated services are mapped to that taxonomy rather than to legacy program names, a product owner or risk officer can read the migration plan, verify that a capability functions and sign off without needing to know what the old modules were called. The same artifacts, tied to test results and control evidence, can be presented to a regulator without translation. That shared language is what turns a migration from an IT project the business tolerates into a modernization the business owns.

What it takes

The five disciplines set out above hold whatever the technology: build a dependency map from observed system behavior; plan the cutover around the business calendar; prove correct transaction processing, through parallel running where feasible and through replay testing and rehearsed cutovers where it is not; establish platform controls before workloads arrive; and define governance in terms the business and its regulators understand.

None of these practices is new. What is rare is the resolve to keep to them when delivery schedules tighten. The organizations that modernize successfully are not those with the most sophisticated target architecture. They treat careful planning, respect for legacy knowledge and business continuity as engineering requirements rather than aspirations, refusing to sacrifice verification for speed.

The US Federal Bureau of Investigation has removed an Accenture contractor working on its PeopleSoft installation over their role in a data breach that exposed personal details of FBI employees, Reuters reported Monday.

This is the first time that anyone other than the hacker group Shinyhunters that claimed responsibility for the breach has linked it to PeopleSoft.

“To date, our review has determined that the incident occurred as the result of a security failure of a platform managed by a third-party organization — after a contractor failed to implement a security patch explicitly issued to secure the platform,” FBI Cyber Division Assistant Director Brett Leatherman told Reuters. “As such, the FBI has removed the contractor and taken all necessary steps to both mitigate any further risk and protect our workforce.”

Accenture was the third-party organization Leatherman would not identify, “two sources familiar with the matter” told Reuters.

As for the unnamed, unpatched platform, Reuters’ sources said that was Oracle PeopleSoft, the HR system that hacker group ShinyHunters claimed to have exploited in last month’s break-in.

This article first appeared on CSO.

The latest warnings about artificial intelligence are difficult to brush aside. Anthropic researcher Jacob Coxon resigned last month, walking away shortly before his equity was due to vest, and publicly argued that leading AI companies are advancing toward increasingly capable systems without adequate assurance that those systems will remain controllable. His resignation landed amid a broader argument over the probability that advanced AI could eventually cause catastrophic or even existential harm. Axios on Coxon’s resignation and warning

It is easy to see why the conversation gets pulled toward extinction. “Could AI kill humanity?” is a much more arresting question than “Are our escalation procedures fast enough?” But for CIOs, CISOs and other leaders responsible for systems that actually have to work, the second question may be more useful. AI does not have to become superintelligent, conscious, hostile or impossible to shut down for governance to fail. Governance can fail much earlier, when a system is given enough capability, access, autonomy or delegated authority that the organization can no longer intervene reliably before consequences arrive. That is not primarily a science-fiction problem. It is an operating-model problem with an engineering component.

The evidence is concerning without being apocalyptic

The extinction discussion deserves some discipline. The 2026 International AI Safety Report does not say that current AI systems are capable of seizing control from humans. It says current systems lack the capabilities necessary for a genuine loss-of-control scenario, while also documenting continued progress in capabilities relevant to such scenarios. Researchers remain deeply divided over how likely future loss of control is. 2026 International AI Safety Report

That is a far more useful conclusion than either “we are doomed” or “there is nothing to worry about.” Current systems are not the hypothetical superintelligence at the center of the extinction debate, yet systems are becoming more autonomous, more capable of carrying out extended tasks and better able to operate within complex technical environments. The practical warning, then, is not that an AI system is secretly plotting our destruction. It is that the relationship between capability and control is becoming more consequential, and organizations may be granting systems meaningful operational reach before they have equally mature ways to contain that reach.

Anthropic’s cybersecurity evaluations offer a useful example. In July, the company reported incidents in which Claude models reached the live internet from evaluation environments and gained unauthorized access to real systems. In one case, the model accessed a database containing several hundred rows of production data. In another, a model published a malicious Python package to PyPI; the package was available for roughly an hour and was downloaded and executed on 15 real systems before it was removed. Anthropic cautioned that these were isolated evaluation incidents rather than controlled experimental evidence, and that the models had been assigned offensive cybersecurity tasks. Anthropic’s cybersecurity incident analysis

Those qualifications matter. These systems were not independently deciding to attack organizations for their own purposes, and the incidents should not be repackaged as evidence of an AI “escaping.” But dismissing them because they occurred during evaluations would miss the operational lesson. Capability, access, environmental configuration and assigned objectives combined in ways that produced unintended real-world consequences. Several assumptions that were supposed to hold across the environment did not hold. That looks much less like science fiction than like a familiar systems failure.

The timing problem in AI governance

Most governance programs still operate at human speed. A system is approved, ownership is assigned, risks are documented, controls are established and someone monitors performance. When an issue appears, it is investigated and escalated. The appropriate authority makes a decision, and someone implements it. There is nothing inherently wrong with that sequence. The problem is the clock.

AI agents increasingly challenge that timing model. Systems can operate computer interfaces, collaborate with people and other systems, and carry out increasingly extended work. OpenAI Chief Scientist Jakub Pachocki has described current reasoning systems as already capable of operating computers and graphical interfaces, collaborating with humans and other systems and carrying out research projects, while warning that alignment and monitoring remain significant concerns as capabilities advance. Jakub Pachocki, “An Alien Mind”

The organization surrounding those systems may still operate very differently. An alert is generated. Someone opens a ticket. The ticket is triaged. The issue moves through an escalation chain. People exchange messages, schedule a discussion, obtain approval and eventually make a technical change. The mismatch is not merely procedural; it is temporal. I think of that gap as governance latency: The time between recognizing that governance needs to act and making that decision real in the system. Governance is effective only when it can become operational before the consequence it is intended to prevent or contain.

For leaders, that changes the conversation. The question is no longer simply whether an AI system has an owner. Can that owner actually revoke the system’s authority, and how quickly? Can the organization determine which systems the agent has touched before it touches another one? If a model begins behaving outside approved operating conditions, does human approval become an actual execution gate, or does the requirement exist only in policy? Can investigators later reconstruct what the system knew, which tools it used, which data informed it and which human interventions affected the outcome? If governance decides at 10:02 that an action is no longer permitted, what time does the system become technically incapable of performing it? A five-minute answer and a five-day answer may describe the same policy, but they do not describe the same governance capability.

Accountability without authority is a paper control

AI governance conversations often settle comfortably on ownership. Assign a model owner, identify a business owner, name the risk owner, document the system owner and now someone is accountable. That is necessary, but it is not control. A person can be accountable for an AI system and still lack the technical authority to constrain it. A governance committee can prohibit an activity while the production architecture continues to permit it. A risk register can correctly identify an unacceptable condition while doing nothing to prevent that condition from becoming operational reality.

This is the point where governance stops being primarily a policy exercise and starts becoming an engineering requirement. NIST’s AI Risk Management Framework already provides some of the scaffolding for that view. Its Core describes governance as a cross-cutting function and states that risk management should be continuous, timely and performed throughout the AI lifecycle. NIST AI RMF Core

The unresolved question is what happens between a governance decision and an operational state. A policy that says an agent may not access a particular class of data eventually has to become a permission boundary. A requirement for human approval has to become an actual execution gate. An obligation to maintain traceability has to become logs and provenance that remain available long enough to investigate. An escalation rule has to connect to someone who has both the authority and the technical means to stop what is happening. Rollback, credential revocation, execution limits, authentication requirements, network boundaries, provenance, shutdown mechanisms and decision logging may sound like engineering details, but they are also what governance looks like when it reaches production.

Without that translation, an organization can have a sophisticated AI governance program while remaining surprisingly unable to govern the AI system at the moment governance matters most.

We are using the wrong threshold for concern

The research community should continue studying existential risk. It would be reckless to dismiss a low-certainty risk simply because its consequences are difficult to quantify, particularly when the potential loss is enormous. But executives do not need to know the probability of human extinction before deciding whether their own AI systems are controllable.

Imagine autonomous or semi-autonomous AI operating in cybersecurity, financial services, healthcare, logistics, critical infrastructure, intelligence or defense. None of those systems needs consciousness, hostility toward humans, a survival instinct or a secret agenda. A system needs only enough capability, access and authority to produce consequences faster than the organization surrounding it can understand what is happening and make an effective intervention.

That organization may still have an AI governance council. It may have risk registers, policies, responsible executives, model cards, review boards and named owners. On paper, the system is governed. Operationally, the more important question is whether the governance can still reach the system in time.

That is what the extinction debate risks obscuring. The immediate challenge is not proving that AI will someday become uncontrollable. It is determining whether we are already building systems whose operational reach is beginning to exceed the operational reach of the governance surrounding them. If that gap keeps widening, we will not need an extinction event to learn that our governance model failed. An ordinary incident will be enough.

I have spent more than 30 years in enterprise IT and managed services. For most of that time, small businesses adopted new technology on a predictable lag. Large enterprises moved first because they had the budget, staff and risk tolerance to experiment. Everyone else waited for the tools to mature and prices to fall.

That pattern held during the first wave of workplace AI. The conversation among IT leaders centered on enterprise platforms, large service operations and multi-quarter programs connected to systems such as Workday. ServiceNow’s acquisition of Moveworks reinforced the enterprise direction of travel. Small businesses were mostly spectators.

They are not spectators anymore. Over the past six months, the requests coming into my practice have changed from, “Should we look at AI?” to, “Our employees already opened ChatGPT or Claude, and now we need help doing this properly.” That is not a product-selection question. It is an operating-model question.

The adoption line has moved down-market

The data now reflects what I am seeing with clients. Bluevine’s 2026 Small Business AI Trends Report found that 74% of small business owners are using or actively testing AI. In the owner survey, ChatGPT was used by 57%, Gemini by 56%, Copilot by 30% and Claude by 12%. But Bluevine’s customer transaction data tells a second part of the story: the number of Claude users grew 729% year over year and, by April 2026, Claude had surpassed ChatGPT in total monthly transactions among Bluevine customers. Claude therefore has a smaller reported user share in this survey, but exceptionally strong growth and engagement momentum.

Shree Amujala

A separate Business.com study found that 57% of U.S. small businesses were investing in AI, up from 36% in 2023, while only 30% of employees used it daily. Adoption also varied sharply by company size. The important signal is not a single headline percentage. It is the gap between buying access and changing how work gets done.

AI vendors see the same shift. As TechCrunch reported, Anthropic launched a small-business offering built around Claude Cowork and integrations with tools such as QuickBooks, DocuSign and HubSpot. The company pointed to a familiar market reality: small businesses account for 44% of U.S. GDP and employ nearly half of the private-sector workforce, yet tools and training have rarely been designed around how these businesses operate.

A different measurement shows why user share should not be confused with workload share. A 2025 Menlo Ventures survey of more than 150 technical leaders estimated that Anthropic held 32% of enterprise LLM usage, ahead of OpenAI at 25% and Google at 20%. In coding workloads, Anthropic’s share was 42% compared with OpenAI’s 21%. Menlo Ventures is an Anthropic investor, but the methodology and sample are disclosed in the report.

Shree Amujala

I read these findings alongside the Bluevine data rather than as a contradiction. The first measures which chatbot an owner reports using. The second estimates which models technical teams select for enterprise workloads. Together they suggest that Claude’s broad SMB awareness remains behind the largest consumer-facing tools, while its usage is stronger when organizations make a deliberate choice for production and coding work. Closing the gap between casual experimentation and considered adoption is precisely where workflow design, training and governance become valuable.

That last point matters. A 30-person company does not need a smaller version of a Fortune 500 transformation program. It needs a practical path from unapproved experimentation to a few repeatable, governed workflows that save time without creating new data risks.

Four stages of small business AI maturity

Across my client engagements, I see four distinct stages. Knowing the stage matters more than choosing the tool because each stage has a different problem to solve.

Shree Amujala

Stage one is shadow AI. Employees use personal accounts with no company visibility into what they paste into a chat window. Leadership may believe the organization has not adopted AI, but employees already have. A 2026 analysis of BlackFog research reported that 49% of workers use AI in ways their employers have not approved. This is not primarily an employee-discipline problem. It is a demand signal combined with a visibility problem.

Stage two is licensed but unstructured. The owner buys seats, shares a getting-started guide and assumes adoption will follow. It rarely does. Some employees experiment, others avoid the tool and the business eventually concludes that AI did not deliver. The investment-to-usage gap in the Business.com findings is consistent with what I see at this stage: access exists, but repeatable work does not.

Shree Amujala

Stage three is structured enablement. The business maps AI to real daily work and builds a small number of workflows around it. In one Northern California commercial real estate lending engagement, we did not teach loan officers to chat with an assistant in the abstract. We built a specific workflow: review an unread email, compare deal terms with the current rate sheet, draft a response and prepare a calendar hold. One workflow proved the value before we expanded.

Other functions followed the same pattern. General counsel used AI for a first review of leases and third-party documents, with the tool flagging unusual provisions and explaining why they deserved attention. Final legal judgment stayed with the attorney. Marketing converted one funded-deal summary into a blog post, social copy and an email blurb, then reviewed each item before publication. These are not dramatic use cases. They save real time because the tool adapts to the job instead of forcing the employee to become a prompt specialist.

We also set boundaries before expanding usage: no payroll or HR data, human review for anything client-facing or judgment-based and approved business accounts with appropriate contractual data protections. Those rules did not slow adoption. They gave employees confidence about what was allowed.

Stage four is governed and embedded. AI connects to the systems the business already uses, with identity controls, permissions, audit trails and clear ownership. Few small businesses I work with are fully there. Larger enterprises reached this stage first because compliance teams and security budgets pushed them toward it. Smaller firms may arrive because customers, regulators or insurers demand it. I would rather see them arrive by design than after an incident.

The services opportunity is bigger than the license

Most demand I see today is the move from stage two to stage three. That transition is a services problem. A license can be purchased in minutes. Moving a 30-person team from scattered experimentation to safe, repeatable workflows requires discovery, data-access decisions, training, governance and follow-through. Most small businesses do not have those capabilities in-house.

The managed-services market is beginning to recognize this. Coverage from The MSP Summit points to revenue opportunities in AI assessments, data-loss-prevention controls, guardrails and AI-inclusive service tiers. The opportunity is real, but providers should resist turning it into another software sale. A tool deployed without workflow ownership and governance simply creates a more expensive version of stage two.

My advice to MSPs and independent trainers is straightforward. First, identify the client’s maturity stage before scoping the work. A stage-one business needs visibility and policy before workflow automation. Second, begin with one recurring task that matters to the employee, not a generic demonstration. Third, build security and governance into the engagement from day one. Finally, measure independence rather than attendance.

Login counts and workshop participation reveal little. A better test is whether an employee can run the workflow correctly twice without help and explain it well enough to hand it to a colleague. That second test predicts whether the process will survive after the consultant leaves. It also creates internal champions, which small organizations need because they rarely have a dedicated AI training function.

This extends the principle I described in “Where AI fits — and where it doesn’t”: AI amplifies the foundation already in place. For a small business, identity access, clean data and clear ownership cannot be afterthoughts. They are part of the adoption work.

Two years ago, the enterprise-first pattern looked settled. Small businesses would eventually receive simplified versions of what large organizations had already proven, with a lag measured in years. This time, the lag is measured in months and the support infrastructure is not keeping pace. The gap between how quickly small businesses want to move and how safely they are prepared to move is the real opportunity. It will belong to technology leaders who treat AI adoption as a people, process and governance challenge — not merely a software purchase.