Anthropic says Trump administration lifted restrictions on some of its most powerful Claude AI models
Artificial intelligence giant Anthropic has announced that the federal government has lifted restrictions on its powerful Claude Fable 5 and Mythos 5 models, ending a weeks-long dispute between the Trump administration and the AI company. The company stated that access would be restored starting Wednesday, following Commerce Secretary Howard Lutnick’s announcement that the models had been approved for use.
Anthropic had recently released the Claude Fable 5 model to the public, with safeguards in place to prevent misuse for cyberattacks or other malicious activities. The company had also introduced the Mythos 5 model, which initially was only available to select major companies for testing purposes. However, shortly after the release of Fable 5, the company was forced to take it down due to export controls issued by the federal government, which required access to be blocked for foreign nationals.
The government’s concerns reportedly centered around a potential “jailbreak” technique that could bypass the model’s guardrails. Anthropic argued that its safeguards were effective and that the vulnerabilities discovered through the alleged jailbreak were relatively simple and replicable with other AI models. The company disagreed with the government’s decision to recall the model and believed that it would hinder innovation in the AI industry.
Last week, reports surfaced that the Commerce Department had begun allowing certain companies and government entities to resume using the Mythos 5 model, easing the strict export controls. The interruption in access to Fable 5 and Mythos 5 highlighted the ongoing debate over the risks associated with advanced AI models and the need for regulation in the industry.
President Trump recently signed an executive order aimed at creating a voluntary review process for private AI models in collaboration with the federal government. The order was delayed as the administration sought to balance U.S. leadership in AI with concerns about potential risks and competition from other countries.
In a separate issue earlier this year, Anthropic clashed with the Pentagon over the use of its AI technology for military purposes. After failing to reach an agreement, the Trump administration ordered federal agencies to stop using Anthropic’s technology for military work, citing supply chain risks. Anthropic sued, and a federal judge blocked the restrictions, describing them as Orwellian and an attempt to cripple the company.
As Anthropic moves forward with the reinstatement of access to its models, the AI industry continues to grapple with questions of regulation, oversight, and the balance between innovation and security. The resolution of the dispute between Anthropic and the federal government marks a significant development in the ongoing conversation surrounding the use of AI technology.
