UK regulators have confirmed that government departments and public bodies must consider freedom of information requests for data about artificial intelligence-generated content. The ruling emerged after New Scientist successfully obtained ChatGPT logs from a government minister under the Freedom of Information Act.
The Information Commissioner's Office (ICO), Britain's independent authority for data protection and freedom of information, clarified that AI systems and their outputs fall within the scope of public disclosure requirements. This means citizens can request to see how government agencies use AI tools, what prompts officials submit, and what content AI systems generate on behalf of the state.
New Scientist's successful request demonstrated the practical application of this principle. The publication obtained records showing how a minister had used OpenAI's ChatGPT for government work, establishing a precedent that such records constitute releasable public information rather than exempt material.
The ruling carries implications for government transparency and accountability. As UK public bodies increasingly adopt AI systems for administrative tasks, policymaking, and service delivery, the confirmation creates a legal framework for public scrutiny. Citizens now have formal grounds to investigate whether AI tools introduce bias, errors, or conflicts of interest into government decision-making.
However, limitations exist. Departments can still withhold information on grounds of national security, commercial sensitivity, or personal privacy. Requests may face practical barriers, as government bodies may argue that extracting AI-related data proves burdensome or technically difficult. The ICO's guidance does not mandate comprehensive AI audits or systemic transparency requirements.
The decision reflects broader global tension between AI adoption and democratic accountability. While governments worldwide embrace AI for efficiency gains, publics increasingly demand visibility into algorithmic systems that affect their services and rights. Britain's approach now positions information access as a check on government AI deployment, though enforcement and institutional capacity to handle such requests remain open questions.