UK and Ukraine strike AI security deal using battlefield data

The UK and Ukraine have agreed a new deal to use Ukrainian battlefield data to train artificial intelligence systems designed to help protect sensitive sites in Britain. The agreement is intended to support security at UK defence locations and critical infrastructure, including railways and energy plants. It is being described as the first arrangement of its kind in the UK.

Sponsored

Under the plan, AI models trained on data from Ukraine will be used to help identify protesters and hostile-state activity around military bases and other protected sites. The scheme will begin with a pilot at a UK defence site, where the system will use AI-optimised sensors in buried fibre-optic cables to detect different kinds of movement. The government has said that, if the technology proves workable, it could later be expanded to airports, prisons, railways and energy infrastructure.

The deal also gives private companies access to data from Ukraine's Avengers AI lab, once approved by the Ministry of Defence on a secure platform. Three British AI companies have already been approved for pilot projects: Sintela, Mind Foundry and Skyral. The data set is said to include four years of military material, including operational drone datasets, strike mission footage and flight profiles, as well as information likely to include Russian sabotage attempts against critical Ukrainian infrastructure.

The agreement comes at a time when governments are looking for new ways to protect critical infrastructure from both physical intrusion and sabotage. In this case, the value of the Ukrainian data lies in the scale and variety of real-world battlefield conditions it captures, which officials believe could improve the performance of security systems beyond what models trained mainly on open-source material can do. That makes the deal significant not only for defence planning, but also for the wider use of AI in national security.

Sponsored

The move also follows a recent security breach at RAF Brize Norton, which was cited in the context of the government's tougher approach to site protection. The activist group Palestine Action said it had targeted RAF Voyager aircraft used for transport and refuelling, while the government has linked the incident to concerns about hostile activity at military sites. The new AI pilot appears aimed at reducing the chance of similar breaches by improving early detection and response.

What remains unclear is how quickly the pilot can be tested, how much of the Ukrainian data will be shared with private firms, and what safeguards will be applied to privacy and oversight. It is also not yet clear how far the system could be expanded if the initial trial succeeds. The next stage will depend on whether the technology can reliably distinguish between routine movement, protest activity and more serious threats.

#UnitedKingdom #Ukraine #artificialintelligence #criticalinfrastructure #defencesecurity