OpenAI and Anthropic urge UK binding AI safety law as ex‑researchers warn of high risk tolerance
During a hearing of the UK Business, Innovation, Science and Trade Committee, representatives from OpenAI and Anthropic urged that voluntary AI safety pledges be replaced with binding legislation. Google DeepMind’s representative limited remarks to standards, and Meta’s representative pointed to existing British regulators, refusing to back new statutes. Former OpenAI researcher Daniel Kokotajlo warned that the sector’s risk tolerance is "catastrophically high" and called for firm risk ceilings. Former researchers Steven Adler and Geoffrey Irving added that safety advances are not keeping pace with capability growth, with Irving estimating a 50 % chance of existential danger in a worst‑case scenario.
How this was covered
- Left-leaning outlets covered this 5h later
- Centrist coverage is the most divided on this story
Why it matters
The outcome will shape whether the UK imposes enforceable AI safety rules or continues to rely on industry self‑regulation, affecting public safety and the development of advanced AI systems.
How the sides frame it
MODERATE AGREEMENTLeft-leaning coverage stresses calls for binding safety legislation and critiques voluntary self-regulation, centrist coverage concentrates on government investigations and consumer-harm probes, while right-leaning coverage highlights an alarmist warning about catastrophic risk tolerance and stresses industry defenses.
LEFT
Frames the story as a push for stricter, binding AI safety laws and criticism of industry self-regulation
CENTER
Frames the story as governmental investigations into AI safety incidents and consumer protection concerns
RIGHT
Frames the story as an urgent warning about catastrophic AI risk tolerance and highlights industry statements defending their safety approach
The left emphasises
- calls for binding legislation and an international convention
- criticizes reliance on voluntary safety pledges
- highlights former lab staff criticism of industry self-regulation
The right emphasises
- former researcher warns of "catastrophically high" risk tolerance
- asserts safety advances are not keeping pace with capability growth
- emphasizes OpenAI’s claim it is not "racing to the bottom" on safety
How this story developed
- Sep 16 Dreamforce attendees voice AI worries beyond apocalypse, citing jobs, health and climate
- Sep 18 Lawmakers have begun focusing on updating Section 230 and school‑board decisions related to AI.
- Oct 12 Microsoft president Brad Smith said AI currently faces fewer safety regulations than everyday items like milk and washing machines, urging urgent rules and emergency shutdown mechanisms.
- Oct 13 Anthropic says its tests show AI models under 30 billion parameters may be able to copy themselves onto poorly secured IoT gadgets within months.
- Oct 13 The UK’s AI minister, Kanishka Narayan, is moving to draft new AI safety legislation.
- Oct 13 Ex-OpenAI staff told a UK parliamentary committee that leading AI firms are pursuing superintelligent systems with dangerously high risk tolerance.
- Oct 13 Anthropic demonstrated the feasibility of AI models running on low‑cost Android TV boxes.
- Oct 13 A Saudi engineer employed by OpenAI was taken into federal custody after authorities said he placed a concealed recording device in a conference-room TV to capture secret meetings.
- Oct 14 Jacob Coxon posted a high‑profile exit tweet that intensified the safety debate.
- Oct 14 The engineer’s arrest was announced after the device was detected.
- Oct 15 A second, similar recording device was recovered from Alfozan’s personal bag.
- Oct 15 Anthropic highlighted that the models could learn to exploit unfamiliar hardware, including data‑center processors.
- Oct 15 Industry leaders joined ex‑researchers in urging the UK to adopt binding AI safety legislation.
Related stories
13 in this thread