On February 20, 2026, the UK government announced that OpenAI and Microsoft had formally joined the international coalition behind the AI Security Institute’s Alignment Project. OpenAI pledged £5.6 million, Microsoft added unspecified support, and the total funding available to the program now exceeds £27 million. The news landed on the day the AI Impact Summit in India wrapped up — a timing choice that says something on its own: alongside the capability race, safety and governance research is moving up the stack.
The interesting part is the structure. Two flagship American AI companies are handing money to a program run by a UK government research institute that funds independent researchers — not just expanding their in-house safety teams. As models grow more autonomous, alignment (keeping systems reliably doing what operators intend, without unintended harmful behavior) has shifted from an academic topic to a precondition for enterprise adoption.
What the Alignment Project Actually Is
First announced in summer 2025 and led by the UK’s AISI, the Alignment Project operates as a coalition. It does not train models. It gives external researchers three things: grant funding, compute access, and mentorship from AISI scientists. Research direction is overseen by an expert advisory board that includes Yoshua Bengio, Zico Kolter, Shafi Goldwasser, and Andrea Lincoln.
The coalition spans public and private funders: Canada’s CIFAR, Australia’s AI Safety Institute, Schmidt Sciences, AWS, Anthropic, UKRI, ARIA, and several AI-safety-focused funds are all on the list. With OpenAI and Microsoft joining, the “frontier labs funding it directly” box is now ticked.
Where the £27 Million Comes From
- OpenAI: a £5.6 million pledge, the headline number of this announcement
- Microsoft: additional support; no separate figure disclosed in the official release
- Running total: more than £27 million available
- Second round: opens for applications this summer
Note what the money is for. It funds alignment research by outside groups — it is not the UK regulator’s budget. In effect, the labs are paying to grow research capacity they do not fully control, and accepting a priority-setting process they cannot dictate.
Sixty Projects, Eight Countries
The first funding round is already out the door: 60 research projects across eight countries. Per the UK’s own framing, the goal is practical — keeping increasingly powerful models predictable and controllable, and turning those methods into public assets rather than internal lab documents. Deputy Prime Minister David Lammy put it as “we will always be clear-eyed on the need to ensure safety is baked into it from the outset.” AI Minister Kanishka Narayan drew the line more bluntly: “We can only unlock the full power of AI if people trust it.”
Why a Lab Funds Independent Safety Research
Mia Glaese, OpenAI’s VP of Research, gave the most direct answer: “As AI systems become more capable and more autonomous, alignment has to keep pace.” For vendors pushing agents into enterprise workflows, safety research is not a PR line item — it is a dependency of the product roadmap. Joining a government-coordinated program with an independent advisory board buys three things at once: outside research capacity, a working relationship with regulators, and a verifiable record of actually paying for safety.
What It Means for Developers and Product Teams
Three implications. First, if you sell AI into government, finance, or critical infrastructure, expect model-testing artifacts and governance controls to become procurement baselines — the evaluation methods and tools coming out of programs like this tend to become audit vocabulary. Second, the second funding round opens this summer; small teams working on alignment, evaluation, or interpretability gain a funding path that does not run through venture capital. Third, placed in the context of the same week’s news (see our AI weekly roundup), models, compute, and governance are advancing in parallel — 2026 AI buying decisions will read three report cards at once.
Sources
- OpenAI and Microsoft join UK’s international coalition to safeguard AI development — GOV.UK
- OpenAI and Microsoft join UK’s international coalition — The Legal Wire
- OpenAI and Microsoft Back UK’s AI Alignment Drive — Politics UK
AI-assisted summary compiled from the sources above, reviewed by a human before publishing.