OpenAI and Anthropic representatives told a UK parliamentary committee that voluntary pledges are insufficient and called for binding AI laws, statutory frameworks, and mandatory incident reporting.
The hearing highlighted divisions among major tech firms regarding self-regulation and government oversight.
The British parliament's Business, Innovation, Science and Trade Committee held a hearing, in which executives from four major artificial intelligence companies, former employees and British civil servants gave evidence. The aim of the proceedings, according to the committee's chairman, Liam Byrne, was to expose the disagreements between the industry's wealthy companies and show that self-regulation is not reliable.
What are Anthropic and OpenAI asking for?
Anthropic's head of Europe, Liam Booth-Smith, told the committee that the rules should be made binding. For his part, OpenAI's vice president Tom Duff Gordon said he supported a legal framework, an international treaty and mandatory reporting of problems and incidents.
In contrast, Owen Larter from Google DeepMind and Rob Sherman from Meta avoided supporting the creation of a new legal framework.
The weaknesses of state control
During the meeting, government officials confirmed that the current control system relies solely on voluntary compliance. Henry de Zoete, head of the UK-based Artificial Intelligence Security Institute, admitted that his organization is entirely dependent on cooperation from companies and foreign governments and has no legal authority to stop the release of a dangerous model.
At the same time, the director of Britain's MI5 intelligence agency, Ken McCallum, warned that artificial intelligence could change the game when it comes to creating new or existing biological weapons. Former OpenAI employees also testified at the hearing, arguing that companies cannot be left to control their own security.
Top comments (0)