Key facts
- A UK parliamentary committee has called for new legislation and an independent oversight body to regulate AI.
- The Joint Committee on Human Rights (JCHR) stated that current AI laws are not fit for purpose.
- The JCHR cited risks including AI-generated sexualized images, unauthorized facial scanning, and inappropriate profiling.
- AI companies like OpenAI and Anthropic have urged the UK government to implement stronger AI regulations.
- The UK government acknowledged the risks of advanced AI and is collaborating internationally.
- Downing Street is exploring targeted interventions for national security risks related to AI.
The UK government is signaling a potential shift towards stricter AI regulation, responding to mounting safety concerns and calls from within parliament and the industry. A cross-party group of MPs and peers, through the Joint Committee on Human Rights (JCHR), has published a report arguing that existing laws are inadequate to address the rapid advancement and potential human rights risks posed by artificial intelligence.
The JCHR's report, which spans 100 pages, highlights issues such as the creation of non-consensual sexualized images, unauthorized facial scanning, and inappropriate use of profiling or biometric data. The committee advocates for a new bill to tackle these threats and the establishment of a "single, independent AI oversight body" on a statutory basis, as the current legal framework is deemed fragmented and insufficient.
Leading AI companies have also voiced support for stronger regulation. OpenAI has urged the UK government to act swiftly, suggesting legislation should focus on the most advanced AI labs. Anthropic, another major AI developer, has called for global coordination to slow AI development, a sentiment echoed by other industry figures. The UK government has stated it takes the risks from advanced AI seriously and is actively monitoring the technology's evolution, working with international partners and AI companies. While a proposed ban on artificial superintelligence (ASI) was rejected, Downing Street indicated it is exploring targeted interventions to address national security risks.