Question to the Department for Science, Innovation & Technology:

To ask the Secretary of State for Science, Innovation and Technology, what steps she is taking to help protect the public from harmful deepfake content.

Under the Online Safety Act, deepfake content is regulated when it is shared on an in-scope service and meets the definition of either illegal content, or content which is harmful to children. The largest user-to-user services will also need to enforce their terms of service including in relation to AI generated content.

We have banned the creation of non-consensual sexual deepfakes, nudification apps, and criminalised AI models which have been optimised to create this content. We have updated the existing laws criminalising ‘paedophile manuals’ to cover the abuse of AI to produce Child Sexual Abuse Material.

We have also introduced new requirements on platforms and search services to remove non-consensual intimate images within 48 hours of a valid report and ensured that senior managers can be held criminally liable when platforms fail to meet their duties. We have also taken powers to bring currently unregulated chatbots within scope of the Online Safety Act’s illegal content duties.

We continue to explore technical solutions to support the transparency of AI-generated content, including through the Deepfake Detection Challenge, the AI Labelling Taskforce and media literacy work.