UK lawmakers and OpenAI push for enforceable AI safety rules

Britain’s Parliament Joint Committee on Human Rights has called for broad artificial intelligence legislation, arguing that voluntary agreements and existing regulators are no longer enough.

The report proposes enforceable transparency and compliance requirements overseen by an independent regulator. It also says AI protections should cover the full supply chain and lifecycle, while responsibility among developers, deployers and users must be clarified.

More than 70 MPs and peers have separately backed calls to prohibit the development and operation of AI. The report also proposes monitoring and control powers. Artificial superintelligence remains hypothetical, however, rather than a current system.

Committee chair Alex Sobel MP said the speed and complexity of AI had left Britain unprepared for potentially serious consequences. He said no country, including the UK, currently has an AI regulatory approach that is fit for purpose.

OpenAI has also argued that voluntary disclosure is insufficient. Its policy manifesto supports mandatory safety requirements for frontier AI companies, independent assessments, and rules for when development should slow or stop.

OpenAI says regulation should focus on the largest and most capable companies rather than apply identical rules to every firm. The source notes that such rules could also benefit wealthy companies by making compliance harder for smaller competitors.

Britain’s AI Security Institute already studies and tests advanced models, including work with OpenAI. The report nevertheless argues that a dedicated statutory regime is needed, while warning that the largest AI companies should not gain extra influence over the rules.