The UK’s AI Security Institute (AISI) has launched the Alignment Project, an international coalition aimed at advancing research on AI alignment and control to ensure artificial intelligence systems behave in ways consistent with human values and goals. The project is backed by over £15 million in funding, offering up to £1 million per research project, along with access to computing resources, venture capital investment, and expert support. Key partners in the initiative include Anthropic, Amazon, and the Canadian AI Safety Institute. Anthropic has committed to contributing compute resources to support the research. The Alignment Project seeks to coordinate global efforts to address urgent challenges in AI safety as AI systems become increasingly capable.