What happened
The US government ordered Anthropic to suspend foreign national access to its Fable 5 and Mythos 5 AI models, citing national security concerns. This action exposes the vulnerability of proprietary, remotely controlled AI services, prompting major European companies like Siemens, Renault, and Orange to accelerate diversification across US, Chinese, and European AI providers. Concurrently, rising token costs, exemplified by Uber exhausting its 2026 budget in four months, compel firms to re-evaluate AI agent deployments and operational expenses.
Why it matters
Access to proprietary frontier AI models now carries explicit sovereign risk, forcing procurement teams and platform engineers to prioritise multi-vendor strategies and on-premises alternatives. This shift reduces reliance on single-source providers, mitigating potential service interruptions and data access restrictions. For financial controllers, escalating token costs necessitate immediate re-evaluation of AI agent deployment economics, shifting focus to cost-per-token metrics and infrastructure optimisation for open-source models. This follows earlier warnings from US officials regarding AI cyber risks to financial institutions.
