News on Medial

US and UK will work together to test AI models for safety threats

The Verge · 1y

The US and UK have agreed to collaborate on monitoring the safety risks of advanced AI models. The agreement includes joint research and at least one safety test. Both countries have emphasized the importance of AI safety, with the US requiring companies to report safety test results and the UK establishing the AI Safety Institute. The partnership aims to accelerate the work of both institutes in addressing risks to national security and society. The possibility of including the European Union in the collaboration is also mentioned, as the EU has already passed its own regulations for AI systems.