#What Happened to Anthropic's AI Models?
The US government recently intervened, leading Anthropic, an artificial intelligence company founded by former OpenAI researchers, to disable its two most advanced models, Fable 5 and Mythos 5. This decision came after new export controls were enacted by the Trump administration on June 12, aimed at addressing national security concerns associated with these models.
The directive from Commerce Secretary Howard Lutnick to Anthropic's CEO resulted in a prohibition that extended to all foreign nationals, including those working within the US. The implications of this were significant: Anthropic chose to stop all access to its models to avoid potential penalties. Consequently, this action affected not only foreign users but also US subscribers and even employees of Anthropic.
#Why Did This Conflict With the Government Develop?
The tension between Anthropic and government authorities has been escalating since February 2026, when federal agencies were mandated to refrain from using Anthropic’s products. This restriction stemmed from the classification of the company as a supply-chain risk. Importantly, the designation was not due to any security breach, but rather a reflection of the company's stringent policies against the military applications of its technologies, particularly for tasks involving surveillance or autonomous weapons.
President Trump labeled Anthropic as a politically driven company earlier this year, which further fueled the conflict.
#How Do Export Controls Affect AI Models?
The recent export controls on AI models represent a new and complex regulatory challenge. Historically, export controls have applied to physical goods such as weapons and military technology. The current situation illustrates an unprecedented application of such controls to software, particularly when delivered through cloud-based APIs.
This latest measure specifically targets Anthropic’s frontier-tier models. By categorizing these AI systems as controlled technologies, the Commerce Department has effectively treated them similarly to dual-use military hardware, rather than as standard consumer software products.
For Anthropic, the compliance implications were severe. In choosing to shut down all access instead of risk penalties by verifying every user's nationality in real time, they opted for a drastic solution. This left US customers without access to paid services and demonstrated the drastic legal and operational ramifications of government actions on tech companies within the AI sector.