Artificial Intelligence: Regulation

Baroness O'Grady of Upper Holloway Excerpts

Monday 10th February 2025 Lords Chamber

[Lord Vallance of Balham (Lab)]

That is precisely the point that I was trying to get to in the last few questions. There is regulation by the existing regulators, all of whom will need to deal with AI, and there is regulation which is covered in the Data (Use and Access) Bill, leaving frontier model control as the unregulated area. That is the area in which we seek to bring in some form of legislation in due course. We want to consult on it; it is a very complicated, fast-moving area, and an important one, and it is why the AI Safety Institute is such an important body in the UK.

[Baroness O'Grady of Upper Holloway (Lab)]

My Lords, does my noble friend the Minister agree that AI has the potential to be a liberating force for workers in terms of repetitive work and so on if workers have strong rights and the gains are shared fairly? Is he aware of the TUC manifesto on AI, and does he agree that workers should have the right to a human review when it comes to recruitment and indeed sackings?

[Lord Vallance of Balham (Lab)]

I completely agree with my noble friend that the aim of AI should be to increase the opportunity for those things that humans can do, and that includes, of course, human-to-human interaction. It is a very important point to consider as this is rolled out, including across the NHS. On automated decision-making, we have been clear that there needs to be human involvement in terms of somebody who knows what they are doing having the opportunity to review a decision and to alter it if necessary.