Google and OpenAI to Provide AI Models to UK Government

During London Tech Week, Prime Minister Rishi Sunak announced that the UK’s AI regulation will involve the opportunity to directly examine the technology of certain companies. Google DeepMind, OpenAI, and Anthropic have committed to granting “early or priority access” to their AI models for research and safety purposes. Sunak hopes that this will enhance the scrutiny of these models and enable the government to identify both the advantages and hazards associated with them.

It is not clear what information the tech companies share with the UK government. We’ve asked Google, OpenAI and Anthropic for comment.

The announcement comes weeks after officials said they were conducting a preliminary assessment of the AI model’s accountability, security, transparency and other ethical concerns. The Finnish Competition and Markets Agency is expected to play a central role. The UK has also committed to spending an initial £100m (about US$125.5m) to set up a Foundation Model Taskforce to develop “sovereign” artificial intelligence aimed at growing the UK economy while minimizing ethical and technical issues.

Industry leaders and experts have called for a temporary halt to the development of artificial intelligence, as the creators are pushing forward without adequate security. Generative AI models such as OpenAI’s GPT-4 and Anthropic’s Claude have been praised for their potential, but they have also raised concerns about inaccuracies, misinformation and abuses such as cheating. The UK move will, in theory, limit these problems and catch problematic models before they’ve done much damage.

This may not give the UK full access to these designs and the code behind them. There is also no guarantee that the government will catch up on all the important issues. However, access can provide relevant insights. If nothing else, the effort promises more transparency for AI at a time when the long-term impact of these systems is not entirely clear.