TL;DR

OpenAI and Microsoft have joined the UK AI Security Institute’s Alignment Project, announced by Deputy PM David Lammy at the AI Impact Summit in India. OpenAI is contributing £5.6 million, bringing total funding to £27 million. The first 60 grants have been awarded to projects across 8 countries.

What Happened

The announcement cements the UK’s position in frontier AI safety research. The Alignment Project, first launched last summer, now has backing from an international coalition that includes the Canadian Institute for Advanced Research, Australia’s AI Safety Institute, Schmidt Sciences, AWS, Anthropic, UK Research and Innovation, and ARIA, alongside the new OpenAI and Microsoft commitments.

Mia Glaese, VP of Research at OpenAI, said the support “complements our internal alignment work and helps strengthen a broader research ecosystem focused on keeping advanced systems reliable and controllable.” She emphasised that the hardest alignment problems “won’t be solved by any one organisation working in isolation.”

The project’s advisory board includes Yoshua Bengio (Université de Montréal), Zico Kolter (Carnegie Mellon), and Shafi Goldwasser (UC Berkeley). It combines grant funding, compute infrastructure access, and academic mentorship from AISI scientists.

Why It Matters

AI Minister Kanishka Narayan framed alignment research as essential to building public trust: “Trust is one of the biggest barriers to AI adoption, and alignment research tackles this head-on.” The government views progress on alignment as the foundation for wider AI adoption across public services, healthcare, and the economy.

The broad coalition — spanning governments, major AI companies, and research institutions — signals growing consensus that alignment is among the most pressing technical challenges as AI systems gain autonomy and capability.

Looking Forward

A second funding round opens this summer. With 4 of the world’s top 10 universities based in the UK, the government is positioning the country as the natural hub for alignment research — a strategic bet that safety leadership can coexist with commercial AI ambition.