OpenAI and Anthropic are co-designing the federal threshold that decides which models face pre-release scrutiny
Following a June 2026 executive order, the two most valuable AI labs are helping design the capability threshold above which a frontier model must undergo government pre-release review. The companies whose models the rule governs are writing the rule — the clearest sign yet that US AI governance is being shaped from inside the industry it regulates.
The structural fact is who holds the pen. A threshold that determines which models face pre-release scrutiny is one of the most consequential levers in AI governance, and having OpenAI and Anthropic co-design it means the incumbents are defining the bar their rivals must clear. Whatever the technical merits, it concentrates rule-making influence in the two firms with the most to gain from where the line lands.
The dynamic cuts two ways. The labs have the most expertise on what a meaningful capability threshold even is, so excluding them would produce a worse rule; including them risks a threshold tuned to incumbent advantage — high enough to wave through their own roadmaps, precise enough to burden a smaller challenger. The design choices inside that threshold are where the real policy is, not in the headline that scrutiny exists.
It also marks the US path diverging sharply from Europe's. As the EU's AI Office begins enforcing a statutory regime written by regulators, the US is building its frontier rule collaboratively with the regulated — a lighter-touch, industry-shaped model whose legitimacy depends entirely on whether the threshold is set in the public interest or the incumbents'.
TechTimes — OpenAI and Anthropic are writing the threshold their rivals must clear for launch → · Anthropic — Newsroom →