London, 27 January 2026, 13:07 (GMT)
- UK government enlists Meta-supported AI experts to develop open-source tools for transport, public safety, and defence
- Meta is investing $1 million to support a year-long fellowship hosted by the Alan Turing Institute
- Later this year, Anthropic will roll out an optional AI assistant aimed at job seekers
Britain announced on Tuesday it has assembled a team of AI experts, with financial backing from Meta, to develop artificial intelligence solutions for transport, public safety, and defence. 1
The move comes as ministers push to deliver quick technology wins within government, following the launch of a separate “CustomerFirst” initiative this month aimed at reducing long waits and repeated form-filling in public services.
This move aligns with a broader effort to harness AI for boosting productivity within government, while ensuring sensitive tasks—like those tied to national security—stay confined to secure systems instead of being sent to external platforms.
The Department for Science, Innovation and Technology announced that Meta is putting in $1 million to back a fresh group of AI fellows. Hosted by the Alan Turing Institute, these fellows will focus on building open-source tools over the coming year. 2
The team will create models to analyze images and video, helping councils prioritize infrastructure repairs more effectively, the department said. They’ll also work on systems capable of running offline or inside secure networks for defense and national security agencies.
Minister Ian Murray declared, “A digital world needs a modern, digital government.” He emphasized plans to “re-wire” key frontline systems across healthcare, policing, and transport.
On Tuesday, another partnership revealed that U.S. AI company Anthropic will join efforts to develop and test an optional assistant for public services. The initiative kicks off with support aimed at job seekers needing career guidance and job hunt assistance. Pip White, Anthropic’s head of UK, Ireland and Northern Europe, described the collaboration as a step toward fulfilling the AI Opportunities Action Plan.
Meta’s role came to light back in July last year, when it revealed plans to work with open-source models like its Llama system. Llama is a “large language model” — a type of software trained on massive datasets that can understand and generate text, as well as process images, audio, and video.
Meta executive Rob Sherman stated, “Meta is proud to help bring top British AI talent into government.” Mark Girolami from Turing added, “AI has huge potential to help us anticipate risks.” 3
The government stated that any tools developed using these models would remain state property, allowing departments to retain sensitive data internally and tailor the technology to their specific requirements.
The pitch carries risks. AI might misinterpret images, reinforce biases in decisions, or deliver confidently incorrect results. Plus, public agencies tend to drag their feet on adopting new tech, even when pilot programs succeed.
The fellows will be working throughout the next year, while the jobseeker assistant pilot is slated to launch later this year. Broader deployment will hinge on how well it performs and the safeguards in place.
