The US and UK governments signed a memorandum of understanding to jointly develop and coordinate advanced AI model testing protocols and safety evaluation frameworks. The US simultaneously noted it is working with 200 companies and entities through its own safety institute to evaluate frontier AI risks. This formalizes government-led testing infrastructure but does not reveal specific technical standards, testing methodologies, or whether this testing produces binding constraints on model release or capability deployment.
331 other tracked developments touching the same claims.