The House of Commons Science, Innovation and Technology Committee heard its first session from Secretary of State for Science, Innovation and Technology, the Rt Hon Michelle Donelan MP, on AI governance and regulation.  This is part of the Committee's ongoing review into governance of AI (which we have written about here).

According to the Committee's summary, the key points were:

• The Secretary of State confirmed that the Government did not plan to introduce AI-specific legislation during the current Parliament, but said that eventually, all nations would have to do so. She argued that the EU AI Act’s likely eventual approach of categorising tools based on the level of risk they pose was “quite a blunt tool”.
• The Government’s response to the AI white paper consultation, and details of how researchers and SMEs can access the recently announced AI Research Resource, will both be published in early 2024.
• The Secretary of State highlighted the model evaluation and testing work being done by the AI Safety Institute but said that establishing a single AI regulator would lead to regulatory overlap and confusion.
• On the open source versus proprietary model debate, the Secretary of State said that the debate should be reframed, to focus instead on the capabilities of models.
• The Secretary of State said that the increasing availability of tools that can produce deepfakes amplified existing threats to democracy, rather than raising new ones. She said that her department was involved in the work of the Government’s Defending Democracy Taskforce as the general election approaches.

Commenting afterwards, our Chair Rt Hon Greg Clark MP said: 

"The Secretary of State was clear in her evidence that the Government does not plan to introduce AI-specific legislation in this Parliament, which in practical terms means before the General Election expected next year. The Government will instead rely on the existing powers of regulators to oversee the deployment of the technology, whilst the AI Safety Institute will undertake technical assessments focused on the safety of future models. 
 

“The Secretary of State told us that the Government’s response to the AI White paper published last March would be made early in the New Year. We look forward to scrutinising the consultation response, particularly what assessment the Government has made of whether the powers of existing regulators are sufficient to respond to the continuing development and deployment of AI.

The video of the session is available here.

You can read our response to the Committee's inquiry into AI governance here and also our response to the UK's White Paper on regulating AI here.

The Secretary of State's comments, and the UK's approach to AI regulation, should be seen in light of the ongoing AI regulatory and governance work at UK and EU level.  In particular, the outcomes of the AI Safety Summit, including the Bletchley Declaration (read more here), proposals in the UK for AI regulation (here), and recent political agreement in the EU on the AI Act (here).  As the Secretary of State said during the session, legislation is only part of government's toolbox and may not be the right one to use; there have been other significant developments in the UK, such as the AI Safety Institute.  Industry should keep an eye out for regulatory developments in parallel with for other government measures intended to support the development and deployment of responsible AI.  

If you would like to discuss how current or future regulations impact what you do with AI, please contact Tom Whittaker, Brian Wong, David Varney, Lucy Pegler, Martin Cook or any other member in our Technology team.