Anthropic Regains Permission to Redeploy Advanced Cybersecurity AI Model to Key US Organizations

In a significant development, Anthropic has announced that it has received permission from the US government to redeploy its Mythos 5 cybersecurity AI model to over 100 major US organizations, including corporations and government agencies. This decision comes after the company suspended access to Mythos and Fable on June 12, following a government directive to restrict access to foreign nationals. The US government had raised concerns that the models could be compromised and used by malicious actors, with reports suggesting a China-linked group had gained access to Mythos 5. According to Anthropic, the government has determined that sufficient safeguards are in place to allow trusted partners to access the model.

The company has been working closely with the government to address concerns and develop protocols for the safe deployment of its models. With this decision, Anthropic will quickly restore access to Mythos 5 for the approved organizations, which operate and defend critical US infrastructure. The company is also in discussions with the government to make its Fable model available again, although a timeline for this has not been announced. This move marks a positive step for Anthropic, which has committed to collaborating with the US government on standards and releases for its AI models.

Source: Engadget