UK must think carefully on legislating for AI
73% of Britons think AI companies face too little regulation and almost four-fifths (79%) have little to no confidence in AI companies to develop the technology responsibly. AI Minister Kanishka Narayan has said that “there is nothing on the table” on AI regulation, however. Our correspondent analyses the current state of play and future prospects for AI legislation. A more certain development is the ICO’s statutory Code on AI and ADM which should be open for consultation in early 2027. While a Code does not have the status of a law, it will be given more legal weight than non-statutory guidance.
As reported by POLITICO, OpenAI is supporting binding UK AI legislation. Its Head of European policy, Tom Duff Gordon, said a while ago that legislation should be binding, but narrowly focused on national security and cybersecurity risks. In the EU, the European Commission is currently taking action following recent loss-of-control incidents, and the AI Office has already requested serious incident reports.
In the daily life of a DPO, AI-generated DSARs are now a fact of life. While requesters may quote case law that does not exist, or wrong legislation, behind most requests there is a genuine need for information or action. It is interesting to see how the courts will cope with inaccurate claims put together without any legal knowledge. Even more worrying is a future where AI-generated information about ourselves will lead to inaccurate decisions, for example in recruitment.
In the area of children’s privacy, Meta’s mammoth US$17 billion settlement in California is also making waves in the UK. Whilst the ICO has been rather subdued in its comments, our correspondent evaluates what this development means for UK-based organisations.
Laura Linkomies
Editor, Privacy Laws & Business
November 2026
Previous |
Contents |
Next |