The Trump administration has dropped restrictions on Anthropic's Mythos and Fable AI models, marking a significant de-escalation in a standoff between the government and the artificial intelligence company.
The move de-escalates a clash between the Trump administration and the company over its cutting-edge artificial intelligence systems. The decision allows Anthropic to restore access to these advanced models.
Anthropic has announced it will begin restoring access to the Fable model starting July 1. This comes after the White House had previously imposed export controls that restricted access to the models for foreign nationals.
The restrictions had effectively limited who could use these powerful AI systems. Anthropic had argued that the order was problematic for its operations and the broader AI ecosystem.
The Trump administration had initially moved against Anthropic's Fable 5 and Mythos 5 models shortly after their release. According to reports, the U.S. government forced Anthropic to shut down these models just three days after their release due to legal restrictions.
The White House ultimately imposed export controls, restricting access to the models for foreign nationals. This created a significant operational challenge for Anthropic, which had invested heavily in developing these cutting-edge systems.
The lifting of restrictions signals a potential shift in how the Trump administration approaches advanced AI regulation. By allowing Anthropic to restore access to the Mythos and Fable models, the government is acknowledging the importance of these technologies for U.S. companies and federal agencies.
OpenAI and Anthropic had previously limited new AI models to Trump-approved customers during a cybersecurity review. This suggests that while restrictions are being lifted, some oversight remains in place.
In our view, this decision reflects a practical understanding that restricting cutting-edge AI models hurts American competitiveness more than it helps national security. The Trump administration appears to have found a middle ground — allowing access while maintaining some level of control.
However, the fact that these models were shut down just days after release raises questions about the regulatory process. Companies need clearer guidelines before investing billions in AI development, not after. The de-escalation is welcome, but the underlying uncertainty remains.
For U.S. companies and researchers, this is good news. They will regain access to some of the most advanced AI systems in the world. But the episode serves as a reminder that AI regulation remains a moving target, and the rules can change quickly.
