Microsoft and OpenAI fund UK AI alignment project
Alignment research guides AI to avoid harmful behaviour, boosting trust, productivity, medical progress, and job creation across communities.
OpenAI and Microsoft have joined the UK’s AI Security Institute, pledging funding to its Alignment Project, an international effort focused on ensuring advanced AI systems are safe, secure, and act as intended.
Their contributions bring total funding to over £27 million, supporting some 60 research projects across eight countries.
AI alignment aims to steer AI systems to behave predictably and prevent unintended or harmful outcomes. The project provides grants, computing resources, and mentorship, boosting public trust in AI while supporting productivity, medical progress, and new job opportunities.
UK Deputy Prime Minister David Lammy and AI Minister Kanishka Narayan highlighted the importance of safe AI adoption. Lammy said strong safety foundations help the UK harness AI’s benefits, while Narayan stressed that public confidence is key to unlocking its full potential.
The Alignment Project operates with a global coalition including the Canadian Institute for Advanced Research, Amazon Web Services, Anthropic, and other partners.
By combining independent research teams, grant funding, and access to infrastructure, the initiative aims to keep increasingly capable AI systems reliable and controllable as they are deployed worldwide.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!