The European Union’s cybersecurity agency was due to meet Anthropic on Thursday, the European Commission confirmed, a meeting arranged before Washington made it considerably more awkward. ENISA, the EU’s cybersecurity body, had been invited to talks with the AI company that were set in motion weeks earlier, well ahead of the US export directive now hanging over them.
The substance of the relationship is access. Anthropic had offered to let ENISA into Project Glasswing, an initiative that lets selected organisations test its Mythos model before wider release, which would make the agency the first European body to get such access.
The arrangement followed months of negotiation between the Commission and Anthropic, including Commission officials travelling to San Francisco in late May to work through terms. The meeting was a continuation of that, not a beginning.
Then the ground shifted. The US Department of Commerce ordered Anthropic to suspend access to its most advanced models, including for foreign nationals, citing the risk they could reach military or intelligence users in countries of concern.
The directive lands directly on the ENISA arrangement: an offer to give a European agency early access to a frontier model, made in good faith, now collides with an American order restricting exactly that kind of access.
