‍

AI model releases are no longer purely private product decisions in the United States. 

The US Commerce Department has set national security review gates for frontier models that pass certain capability thresholds, meaning major model launches now require government review before they reach the public. 

What Triggered the Change

Major releases such as GPT-5.6 and Claude Fable 5 now need government review before launch, marking a shift in how frontier AI reaches the market. 

The move reflects growing government concern that the most capable models carry national security implications beyond ordinary product risk. 

The scope is expanding beyond software too:

• The FCC has added foreign-made humanoid and quadruped robots to its national security review list 
• Capability thresholds, rather than company size or funding, appear to determine which models trigger review
• This adds a new, non-optional step to launch timelines for frontier AI labs

Why It Matters for Enterprises

For businesses building on top of frontier models, this signals that release timelines for the most powerful models may become less predictable, and that government oversight of AI is shifting from guidance to gatekeeping at the point of launch.

CloudEagle helps enterprises track which underlying models and vendors power their AI stack, so if a review or delay affects a specific model, teams already know exactly where their exposure sits.

‍

Your AI Stack Is Bigger Than You Think

Find every hidden tool.

Get The Guide