Anthropic recently faced an unprecedented request from the White House, likely in connection with the Trump administration’s efforts to restrict foreign access to certain Anthropic models due to identified potential jailbreak vulnerabilities. This action came after senior officials from the Department of Commerce engaged in discussions with Anthropic leadership, emphasizing the government’s proactive approach to AI security.

Anthropic: Anthropic is an AI research company known for developing advanced language models including the Claude series as well as newer releases such as Fable and Mythos. The company focuses on building reliable and controllable AI systems for enterprise and research use. It is currently navigating a direct request from the US government regarding restrictions on access to its latest models.
White House: The White House serves as the executive office of the President of the United States and oversees federal policy on technology, national security, and innovation. Under President Donald Trump, it has engaged directly with leading AI firms on matters of model safety and export controls. The office issued an unprecedented directive to Anthropic concerning access restrictions for its advanced AI systems.

`json
{
“AI Security”: “Anthropic is facing significant security challenges following a request from the White House regarding their models.”,
“Government Action”: “The White House has engaged with Anthropic leadership on certain directives.”
}
`