This is the new MLex platform. Existing customers should continue to use the existing MLex platform until migrated.
For any queries, please contact Customer Services or your Account Manager. Dismiss

AI capabilities improving fast, models still vulnerable, UK testing agency says

( December 18, 2025, 13:33 GMT | Official Statement) -- MLex Summary: Frontier AI models are improving across all areas tested by the UK AI Security Institute, the agency's first trend report showed on Thursday. After testing more than 30 frontier models, the institute said they had improved rapidly in cyber tasks – both offensive and defensive – and now far surpassed PhD-level experts in aspects of chemistry and biology. All models tested were still vulnerable to jailbreaks – techniques to override safeguards – but it took the testers 40 times longer to break in than two years ago, it said. Models’ capabilities for evading human control were also improving, it added. Technology ministry press release attached. The AISI summary and full report are attached:...

Prepare for tomorrow’s regulatory change, today

MLex identifies risk to business wherever it emerges, with specialist reporters across the globe providing exclusive news and deep-dive analysis on the proposals, probes, enforcement actions and rulings that matter to your organization and clients, now and in the longer term.


Know what others in the room don’t, with features including:

  • Daily newsletters for Antitrust, M&A, Trade, Data Privacy & Security, Technology, AI and more
  • Custom alerts on specific filters including geographies, industries, topics and companies to suit your practice needs
  • Predictive analysis from expert journalists across North America, the UK and Europe, Latin America and Asia-Pacific
  • Curated case files bringing together news, analysis and source documents in a single timeline

Experience MLex today with a 14-day free trial.

Start Free Trial

Already a subscriber? Click here to login

Documents

Related Sections