Trump Administration Places Export Controls on Anthropic and OpenAI Models, Decentralized AI Tokens Rally

What the Government Did

On June 13, 2026, the Commerce Department imposed export controls on Anthropic's Fable 5 and Mythos 5 models, forcing the company to temporarily suspend global user access, according to Crypto Briefing. Within two weeks, OpenAI announced that its GPT-5.6 Sol model would be available only to customers explicitly approved by the Trump administration while a cybersecurity review was underway.

Two of the largest private AI companies in the world were effectively put on a government leash inside the same month.

Commerce Secretary Howard Lutnick has been the central regulatory figure steering these decisions. A June 2026 executive order established the framework behind both moves: a voluntary review process giving the federal government up to 30 days to evaluate advanced AI models before they reach the public.

Anthropic's Partial Recovery

Anthropicā€™s CEO Dario Amodei negotiated a partial resolution. By late June, the company had restored access for trusted partners after discussions with U.S. officials, per Crypto Briefing. The broader export restriction remained, but Anthropic was NOT in a full lockout by the time those talks concluded.

What "trusted partners" means in practice, and how the approval process works, has not been detailed publicly.

The Crypto Response

Bittensor's TAO token surged approximately 30% in the window following the Anthropic restrictions, according to Crypto Briefing. Internet Computer's ICP token and Venice Token also moved higher in the same period. The pattern suggests a deliberate rotation toward AI projects built on permissionless, decentralized architectures. Platforms built this way cannot be restricted with an export-control order.

If Washington can throttle access to Anthropic and OpenAI, those models carry regulatory risk that open and decentralized alternatives do not.

The Real Limits of That Trade

Critics of the TAO-and-ICP rotation point out that decentralized AI networks are still early-stage relative to what Anthropic and OpenAI have built. The models running on these platforms are generally less capable. The infrastructure is less mature. The user experience is rougher. A 30% token price jump does not close a multi-year technical gap overnight, and token prices have a long history of running far ahead of actual adoption.

Capital rotating into decentralized AI because of regulatory tailwind is not the same thing as developers actually deploying workloads at scale on those networks. Whether projects like Bittensor can convert this moment into real users running real workloads, rather than just speculative token demand, is an open question that a single price rally cannot answer.

Why This Matters Beyond Crypto

The export-control move on Anthropic and OpenAI is a significant shift in how the U.S. government treats advanced AI. Prior to this, the dominant federal posture was voluntary safety commitments and light-touch guidance. A mandatory pre-release review window with export controls as the enforcement mechanism is a different regime entirely.

From a national-security standpoint, the administration's position has an obvious rationale. The most capable AI models represent potential dual-use technology, and restricting their global availability during a cybersecurity review is consistent with how the U.S. has handled semiconductor exports and other sensitive technologies. That argument is strongest when applied to foreign adversaries; it gets harder to defend when it disrupts access for allied nations and commercial customers with no security concerns.

From a market standpoint, the immediate consequence is that two of the most capable AI model families on the planet became less accessible to the global developer community at the same time. Where developers route around that restriction—toward open-source models, toward decentralized networks, or toward non-U.S. alternatives—has real implications for where AI talent and investment concentrates over the next several years.

The Unresolved Question

The June 2026 executive order's 30-day review framework was described as voluntary, yet the Anthropic export controls and the OpenAI access restriction both came with mandatory force. Whether the administration intends to apply this review framework broadly across the AI industry, or whether Anthropic and OpenAI were targeted specifically due to their model capabilities, has not been explained publicly. If additional companies receive similar restrictions, the rotation into decentralized AI that started with TAO's 30% rally could accelerate considerably. If these controls remain narrow and temporary, the move may prove to be a one-time disruption rather than a structural shift in how frontier AI gets built and distributed.