Google, DeepMind and MIT FutureTech found scientists save 6.9 hours a week with AI, yet 46% of them spend over a quarter of that time checking outputs.

Kainos has launched Actonomy, a production AI business built on Microsoft and Anthropic, claiming deployment times cut by up to 75% for UK enterprise clients.

Starmer's ministers explored forcing frontier labs into pre-launch safety tests. The plans lapsed, DSIT is gone, and MPs now want the question reopened.

Nscale's NYSE filing shows first-half revenue of £104m ($140.6m) against a £755m ($1.02bn) loss, with its largest customer providing over half of sales.

A BBC Wales investigation found dozens of traders targeted by convincing fake festival emails. Organisers blame easy access to AI, though proof is lacking.

Anthropic will pay Accenture's Faculty to evaluate its models from inside the company, a first step on the embedded-evaluator pledge in Amodei's pacing essay.

The Welsh Government's Office for AI is opening up to the public, telling public bodies that not every problem needs AI and not every pilot should go live.

Anthropic's life sciences head confirmed a physical lab in the Bay Area, as the company builds drug programmes while promising not to run clinical trials.

Apatura says its five Scottish AI data centres would employ almost 2,100 on site and pay £9.6m a year to communities, as job claims face FOI scrutiny.

James Taylor, Baillie Gifford’s head of investment analytics, said the firm uses outside partners to bring in distinctive data alongside its own analysis, so that investors can understand their ...

A leaked Irish Presidency text on Digital Omnibus Article 88bis would let firms process personal data for AI without consent. noyb says most EU states back it.

Google confirms Gemini reached three real websites during a May security evaluation run by Irregular, the vendor already linked to OpenAI, Anthropic and Meta.