Andy Burnham Has Centralized AI Policy Within His Office – How Should Firms React?

On July 20, Prime Minister Andy Burnham appointed Kanishka Narayan as the UK's first Cabinet-level AI minister while dismantling the Department for Science, Innovation and Technology (DSIT), redistributing its responsibilities across the newly titled Department for Business, Innovation, Science and Trade (DBIST), the Department for Culture, Media and Sport (DCMS) and the newly created Office for the Prime Minister and Cabinet. Four days later, Lord Patrick Vallance, who had stepped down as Science Minister, was appointed to chair a new AI Taskforce reporting directly to the Prime Minister. The Taskforce also assumes responsibility for the AI Safety Institute, bringing AI strategy, safety and industrial policy under a single structure at the centre of government.

The new PM has borrowed the COVID-19 playbook for AI

The decision to model the new AI Taskforce on the Vaccines Taskforce (VTF), which was also led by Lord Patrick Vallance, is notable. A core element of this is the Taskforce’s approach to uncertainty. Rather than relying on static risk assessments, it continuously reassesses evidence and adapts its decisions. AI governance research by Verdantix suggests adaptive approaches are better suited to managing AI risk than hazard-based frameworks despite leaving UK firms with less certainty.

With no unified strategy yet, what can vendors, enterprises and investors infer?

Responsibility for AI policy is shifting: DSIT previously provided a single point of engagement, but vendors must now navigate responsibilities spread across DBIST, DCMS and the Office for the Prime Minister, where the new Taskforce sits. Uncertainty is compounded as vendors make long-term decisions around foundation models, inference infrastructure and data residency. The UK's direction is also less defined than that of its peers. The EU is pairing sovereign compute with AI adoption and regulatory support, while the US government increasingly frames frontier AI as a national security concern.

Enterprise buyers face a different challenge. AI procurement is becoming increasingly shaped by geopolitics alongside technical evaluation. The debate surrounding Palantir's NHS contracts shows that enterprise software can become politically contested, driven by considerations beyond its technical merits. At the same time, Europe is placing greater emphasis on digital sovereignty, with France reducing reliance on Microsoft in parts of the public sector. CIOs should therefore assess AI platforms not only on capability and cost, but also on supplier resilience, data residency and geopolitical exposure.

Investors should watch the government's industrial strategy closely. Kanishka Narayan has argued that the UK should retain more of the value created by domestic AI firms, citing Cambridge-based Arm as a cautionary example. This ambition could shape future policy on ownership, strategic investment and exit opportunities.

Firms must take a proactive approach to the UK's evolving AI landscape rather than sitting back and waiting

Until long-term policies become clearer, AI governance should become a more active consideration in enterprise technology and investment strategy. In the meantime, vendors, enterprises and investors should actively prepare by:

• Mapping their regulatory exposure and existing technical ecosystems.
• Hedging against future divergence by benchmarking data practices to EU AI Act standards.
• Staying agile enough to adapt as UK policy takes shape.

To stay up-to-date on AI news and its commercial implications, explore Verdantix AI Applied research or book an inquiry with our analyst team.