A cross-party select committee has urged the government to introduce a dedicated Artificial Intelligence Bill, arguing that current legislation is inadequate to address serious threats to human rights. The Joint Committee on Human Rights (JCHR) published a comprehensive report warning that the UK, like the rest of the world, lacks a regulatory framework fit for purpose.
Alex Sobel MP, the committee’s chair, emphasized the urgency of the situation, stating that no jurisdiction globally currently possesses an AI regulatory approach that adequately handles the scale and severity of potential harms. He noted that the technology is evolving with such speed and complexity that society is presently unprepared to manage its consequences, which could prove dire.
The 100-page report highlights several specific human rights violations facilitated by AI, including the creation of non-consensual sexualized images of women and girls and the unauthorized scanning of facial data. The committee also pointed to the risk of algorithmic bias embedded in systems trained on internet-scraped datasets containing racist or sexist material, as well as the potential for AI to generate and disseminate misinformation.
To combat these issues, the JCHR is advocating for the establishment of a single, independent, statutory oversight body. The proposed bill would implement a risk-based regulatory regime, imposing stricter obligations on high-risk AI systems and models. Responsibilities would cover the entire AI lifecycle, ensuring that designers, developers, and users within the supply chain are held accountable.
The committee recommended that certain applications be banned outright if they are incompatible with human rights, specifically citing subliminal techniques and the inappropriate use of profiling or biometric data. The report argues that the existing legal framework is fragmented, creating gaps in protection that leave individuals vulnerable.
The call for legislative action comes amid heightened anxiety within the tech industry. Recently, Dario Amodei, CEO of AI firm Anthropic, proposed a plan for global regulation and independent monitoring of AI development, urging a slowdown in progress. His stance was supported by rivals Sam Altman of OpenAI and Elon Musk. Additionally, Jacob Coxon, a former Anthropic researcher, told the BBC that staff were
I’m skeptical about adding another bureaucratic layer. Wouldn’t a dedicated oversight body just slow down innovation further without actually solving the core issues?
Finally, someone is taking this seriously. The current patchwork of laws leaves us completely exposed to algorithmic bias and privacy violations.