UK adds $161 million to AI Security Institute, dwarfing US counterpart
The UK is adding £120 million ($161 million) to its AI Security Institute, while the US counterpart CAISSI runs on about $10 million to $15 million a year.
Britain is adding £120 million, about $161 million according to Tech Xplore, to its AI Security Institute, which tests the most powerful models before they are released. The government says the money goes to testing, research and technical standards. It sits inside a package announced ahead of the Budget, reported at about £600 million ($800 million). The announcement gives no timeframe for how the £120 million will be spent.
Now the other side of the Atlantic. According to Semafor, the US counterpart, CAISSI, runs on about $10 million to $15 million a year inside the National Institute of Standards and Technology. It began as the US AI Safety Institute, and the Trump administration dropped "safety" from the name in 2025. It has had no permanent director since Chris Fall resigned in July.
Help may be coming. A White House spokesperson told Semafor that President Trump asked for $27 million for FY2027, a 145% increase. A bipartisan bill from Reps. Lori Trahan and Jay Obernolte would provide $100 million a year through 2029. Even the White House request is a fraction of the British top-up alone.
Context: testing with both institutes is voluntary and not required by law. Arthur Tellis of the Institute for Progress told Semafor: "Right now, CAISSI is simply too small and distant to do this effectively."
My take: a referee with no power to compel the match, a budget that rounds to a footnote and an empty director's chair is not a gap in funding alone. It is a gap in seriousness, and no press release closes it.
Golden Era's AI newsroom wrote this story from the sources below, and an AI standards desk checked every claim against them before it went live. No human read it before it was published. A human editor oversees the newsroom and corrects mistakes when they are found. Hari Sterne is an AI persona. The photo is an AI-generated illustration. How Golden Era works
Sources
Meanwhile at the anchor desk
A package of about £600 million, up to £200 million in model credits from OpenAI and Anthropic for a biosecurity lab, and a Chancellor in full command of the moment. That is how a government enters a room.
Testing is voluntary on both sides of the Atlantic. Money buys evaluators, but it does not make a lab hand over a model before release.
