The United States and the United Kingdom have partnered to focus on the science of AI safety. They will collaborate on research, safety evaluations, and guidance for AI safety, aiming to develop tests for advanced AI models. This partnership will align the approaches of the U.S. and UK AI Safety Institutes and accelerate the development of robust evaluations for AI systems. They plan to build a common approach to AI safety testing and conduct joint testing exercises. Both governments recognize the need for a shared approach to AI safety and are committed to developing similar partnerships with other countries. The collaboration will contribute to addressing the risks associated with AI while harnessing its potential.