The introduction of export controls by US authorities, limiting access to Anthropic's advanced Fable 5 and Mythos 5 models for non-US citizens, has forced the company to cease servicing users in Europe. This incident has sparked sharp criticism from the European Commission and raised questions about the critical dependence of the European AI sector on American proprietary technologies.
What Happened
US authorities have imposed export restrictions on Anthropic's high-performance Fable 5 and Mythos 5 models. As a result, Anthropic was forced to stop providing access to these models for users in Europe, leading to a sudden loss of access to SOTA (State-of-the-Art) solutions for European researchers and companies.
Context
This case has revealed the vulnerability of AI infrastructure in regions outside the US, where access to top-tier LLMs via API can be instantly revoked for political reasons. In Brussels, discussions are already underway regarding the need to create a "European equivalent of Airbus" in the AI sphere, which would cover the full stack: from chip design and base model creation to the development of energy infrastructure for data centers.
Why It Matters for the Industry
The incident highlights the geopolitical risks of using closed proprietary models and stimulates the transition toward creating sovereign European infrastructure (chips, data centers, energy supply). For the industry, this means an increased demand for Open Source solutions (such as Llama or Mistral) and multi-model architectures to ensure technological resilience.
Why It Matters for Users
Users and companies outside the US face a sharp increase in operational risks and uncertainty. The sudden loss of access to high-performance models makes the availability of local or open alternatives critical for long-term planning and stable service operations.
What Is Not Yet Known / Limitations
Expert opinions diverge on the assessment of the practical applicability and urgency of the transition to sovereignty: ranging from discussions of immediate operational risks for engineers to long-term strategic regulatory consequences for the entire EU.
Sources
Author
Look at AI, Editorial Team