Concern is mounting that the UK government is not up to speed in acting on AI risks, as experts warn of existential threats and political attention wavers. The issue has resurfaced after two weeks of stark warnings, including a claim that there is a greater than 10% chance AI could “kill all humans” and King Charles’s caution about its destructive capabilities. Yet, despite these alarms, critics fear the government’s focus has shifted elsewhere.
Political Turmoil Stalls AI Safety Legislation
Towards the end of Keir Starmer’s time in office, senior ministers, alarmed by rapid AI developments, began drafting plans for a new AI safety law. They ordered a review of existing legislation to identify available powers and explored forcing advanced technology companies to submit products for safety testing before launch. However, as political chaos engulfed the former prime minister’s fight to stay in office, these plans fell by the wayside.
Now, under Andy Burnham’s leadership, the government’s attention has turned again to AI. Burnham has said relatively little about the technology and, in one of his first acts, scrapped the department leading the legislative work. His focus is on immediate domestic problems like balancing the first budget and reducing the cost of living. Some fear this means the threat posed by AI has dropped off the government’s radar.
The Urgent Need for AI Safety Measures
“The first duty of government is to keep people safe,” said one expert, echoing widespread concerns. The UK’s current approach to AI regulation is fragmented, with no dedicated safety law. This leaves gaps in oversight, especially for advanced AI systems that could pose significant risks.
Without mandatory safety testing, companies can release powerful AI tools without independent scrutiny. This contrasts with other sectors like pharmaceuticals, where rigorous testing is required before public use. As AI becomes more capable, the stakes grow higher.

Key Takeaways
- UK AI safety legislation has been stalled by political instability.
- Experts warn of existential risks from advanced AI.
- The government’s focus on domestic issues may delay action.
- Mandatory safety testing for AI products is not yet in place.
- Public figures like King Charles have raised alarms about AI’s destructive potential.
Comparing AI Safety Approaches
To understand the UK’s position, consider how it stacks up against other regions. The table below highlights key differences in AI safety measures.
| Region | AI Safety Law | Mandatory Testing | Regulatory Body |
|---|---|---|---|
| UK | No dedicated law | No | Fragmented |
| EU | AI Act (proposed) | Yes for high-risk | European AI Board |
| US | No federal law | No | Sector-specific |
| China | Several regulations | Yes for some | Multiple agencies |
The Path Forward for UK AI Policy
For the UK to effectively manage AI risks, a comprehensive safety framework is essential. This could include a new AI safety law, mandatory pre-deployment testing for advanced models, and a dedicated regulatory body. Such measures would help ensure that AI development aligns with public safety and ethical standards.
However, political will is crucial. With competing priorities, AI safety must not be sidelined. As the technology evolves, the window for proactive regulation narrows. The government must act decisively to protect citizens and maintain the UK’s position in the global AI landscape.