Most people are treating this like a routine outage.

It is not.
On June 12, the US government issued an export control directive ordering Anthropic to suspend access to its two newest models, Claude Fable 5 and Claude Mythos 5, citing national security concerns.
Here is the part that should stop you mid scroll.
The directive forces Anthropic to disable both models for all users worldwide, not just the foreign nationals the order was technically aimed at.
That includes Anthropic's own non US employees working from offices in the United States.
This is the first time the US government has pulled a publicly deployed frontier AI model offline by force.
Here is the timeline.
Both models had only launched three days earlier, on June 9.
The directive arrived at 5:21pm Eastern.
Anthropic complied within hours.
Access to other models, including Claude Opus 4.8, was not affected.
Now here is what most people are getting wrong.
The government's concern was a method for bypassing or jailbreaking Fable 5's safeguards.
Anthropic says the jailbreak is narrow, and similar capability already exists in models like GPT 5.5.
The legal basis is the deemed export doctrine, a rule built decades ago for semiconductors, now being applied to AI.
Why this matters beyond one company.
Chinese open source models went from roughly 1.2 percent of Hugging Face downloads to around 30 percent in a single year.
The US is now treating frontier AI as a strategic asset, not a normal product.
If a government can pull a frontier model offline overnight, every company building on a single AI provider just got a lesson in concentration risk.
The next fight will not be about jailbreaks.
It will be about who decides which AI the world gets to use.
Tell me I am wrong.