FrontierGuard, explained for beginners

FrontierGuard shows in seconds whether an AI developer likely falls under the rules written for the biggest AI labs.

What it is

New AI laws include special duties for the developers of the largest, most capable models. Those rules usually apply only above certain thresholds: how much was spent on computing, how much revenue the company makes, how much computation went into training.

FrontierGuard takes those numbers and shows, test by test, whether a developer likely crosses the thresholds. It is a screening tool to decide whether to ask a lawyer, not a legal opinion.

An everyday comparison

A tax-bracket check for AI developers.

How it works, in 4 steps

1. Enter compute spend, revenue, training compute and whether a frontier model was trained.
2. Each threshold test is checked separately.
3. Answer: likely covered or likely not, with each test shown.
4. Take the result to counsel.

Try it now

Open the three live demo tools → Every button sends a real request to the live product and shows the real answer.

See it on the homepage: FrontierGuard section →

Who it is for

Model developers and their investors and advisers. A narrow audience: most companies are far below the thresholds.

FIRME AI CORPORATION · CAIN-42 is a Byzantine trust fabric. Numbers marked "our own check" were run by us, not by an independent party. Nothing on these pages is a certification.