US issued order on Anthropic's models — a new era of AI oversight begins

The U.S. government has issued an official order directly concerning specific Anthropic models — the developer of Claude. Bloomberg calls this document a signal of the advent of a new era in artificial intelligence regulation: for the first time, American federal authorities are targeting an individual developer rather than the technology industry as a whole.

What Happened

The document reported by Bloomberg introduces a new regime of government oversight over Anthropic's AI models. At the center of the reporting is a project with the codename Mythos, whose nature remains subject to dispute: either it is a classified version of Claude created for the needs of federal agencies, or it is a separate mechanism for joint management of powerful AI systems under government control. Bloomberg discloses the specific conditions of the order only partially. However, the general conclusion is unambiguous: a precedent has been established. The American government now has tools for direct regulation of commercial AI systems of a specific developer — not merely recommendations, but a binding official act.

Why Anthropic

The choice of Anthropic as the subject of the first such order is not accidental. The company occupies a special place in the AI industry:

• Develops some of the most powerful frontier models — the Claude line and its successors
• Consistently positions itself as a "responsible AI lab," which creates heightened expectations from regulators
• Has received strategic investments from Google and Amazon, making it systemically important to the American technology economy
• Actively participates in government AI safety initiatives — the National AI Safety Institute and related programs

This combination of technological weight, corporate connections, and public position on safety may have made Anthropic a convenient "pilot case" for testing out a new regulatory mechanism.

The Mythos Enigma

The key element of the reporting is Mythos. Bloomberg directly calls this story "convoluted," without fully disclosing details. Depending on the interpretation, the discussion could concern:

• A closed version of Claude, developed at the request of U.S. military or intelligence agencies
• An internal Anthropic program for monitoring and restricting potentially dangerous capabilities of its models
• A joint American-corporate mechanism for real-time oversight of frontier AI systems

Regardless of what exactly lies behind the codename, the very fact of its appearance in the context of a federal document is fundamental: AI regulation in the United States is moving beyond abstract declarations.

"This order signals the advent of a new era in AI control," —

Bloomberg Tech.

What It Means

The United States has long avoided strict federal AI regulation, preferring voluntary agreements with developers and advisory standards. This order changes the logic: the government is beginning to work directly with specific systems and specific companies, rather than with the industry as a whole. For Anthropic, OpenAI, Google DeepMind and other frontier AI leaders, this is a clear signal — the era of independent determination of safety rules is coming to an end, and it is being replaced by formal government oversight.