Trump could unlock Anthropic's most powerful model: the regulatory shift that redefines government control over frontier AI

By Zendoric · June 28, 2026.

According to information gathered by the Chinese financial platform Futunn, the Trump administration is reportedly close to lifting the restrictions that until now limited the deployment or export of Anthropic's most powerful model, designated in the source as 'Fable 5,' with an unblocking horizon of roughly a week. The news comes from a Chinese financial outlet and the model's designation could be an artifact of automated translation from English into Chinese and back; for that reason, that detail should be treated with caution until Anthropic or government sources confirm it directly.

What is significant is the regulatory backdrop: the very existence of government restrictions on a commercial Anthropic model is, in itself, a telling indicator of the current state of AI policy in the United States. Under the Biden administration, AI diffusion controls were introduced affecting both chips and, in a more nuanced way, access to frontier models in certain markets or usage contexts. The debate over whether those restrictions foster safety or simply penalize U.S. companies against less scrupulous competitors has never been fully settled.

The Trump administration's stance on AI regulation has been, broadly, more oriented toward competitiveness and reducing barriers than toward the precautionary approach of the previous period. If the unblocking is confirmed, it would fit that line: prioritizing that the most advanced AI tools be available —to companies, allies or strategic markets— over maintaining restrictions whose real effectiveness is debatable when open-source models proliferate globally anyway.

For Anthropic, whose public position has always combined responsible-safety messaging with very concrete commercial ambitions, lifting restrictions on its most capable model would open up monetization and deployment opportunities that until now would have been administratively blocked. As sector context, Anthropic's competitors —OpenAI, Google DeepMind, xAI— have navigated this same regulatory maze with varying degrees of government exposure and preferential-access arrangements.

The most interesting reading is not the unblocking itself, but what it implies about who, ultimately, holds the key to the world's most powerful models. That a change of administration can activate or deactivate access to private technology of this magnitude lays bare a structural tension: frontier AI operates at the crossroads of commercial innovation and state strategic asset, and that crossroads still has no clear or stable rules. Any company or government planning its AI strategy without taking that regulatory dimension into account is ignoring one of the most real risks of the current ecosystem.

Pending official confirmation of both the specific model affected and the exact terms of the possible unblocking, this news warrants direct monitoring of primary sources at Anthropic and the U.S. federal government before drawing operational conclusions.