White House Raises Alarms Over Anthropic's High-Risk AI Model Mythos

Earlier this year, Anthropic unveiled a preview version of Mythos, an upcoming AI model it claimed was too dangerous for widespread public release. The company granted early access to approximately 50 businesses and organizations, asserting that Mythos could help them prepare for potential cybersecurity threats.

According to Anthropic’s researchers, Mythos is capable of identifying and exploiting large-scale cybersecurity vulnerabilities, potentially giving hackers a significant advantage. Last week, the company disclosed it was investigating a potential unauthorized access to the Mythos model.

These risks have now alarmed even the White House. Wall Street Journal reports that officials under the Trump administration oppose Anthropic’s plan to expand Mythos access to roughly 70 additional companies and organizations, citing security concerns.

One source told the newspaper that Mythos could consume excessive computing resources, potentially hindering the US government’s ability to use it effectively—a claim Anthropic has since denied.

Anthropic’s Strained Relationship with the Department of Defense

This development adds another layer to Anthropic’s already contentious relationship with the Department of Defense. In late February, Anthropic CEO Dario Amodei refused to allow the military to use its AI models for autonomous weaponry or mass surveillance of US citizens. This refusal reportedly angered Defense Secretary Pete Hegseth and President Donald Trump.

Ironically, Anthropic remains classified as a supply chain risk by the White House, a label officials applied seemingly in retaliation. This designation forces defense contractors to sever ties with Anthropic, despite reports that the US military relied on its AI models during operations in Iran.

During a CNBC interview today, Department of Defense CTO Emil Michael described Mythos as a “separate national security moment,” suggesting its risks are distinct from Anthropic’s broader classification.

Ongoing Negotiations and Uncertain Future

An Anthropic spokesperson told the Wall Street Journal that the company is in productive discussions with the government regarding the phased rollout of Mythos access to more organizations.

For now, the full capabilities and risks of Mythos remain uncertain. White House AI advisor David Sacks recently tweeted, “A growing number of people are wondering if Anthropic is the AI industry’s ‘boy who cried wolf.’ If Mythos-related threats don’t materialize, the company will have a serious credibility problem.”