GreenSquare AI

Test the method.
Publish the boundary.

We study whether structured decision support improves professional judgement, and state plainly what the evidence cannot yet establish.

The structured process is where the tested change showed up.

In our initial preregistered benchmark, the loaded condition matched or exceeded an informed baseline on the measured judgement outcomes.

That is evidence for a structured method under the tested conditions. It is not proof of universal performance.

Read the full benchmark

Under the tested conditions, the structured method improved how the decision was worked.

Supported: the loaded condition matched or exceeded the informed baseline on the measured judgement outcomes.

Not established: performance across every decision type, organisation, user population or model environment.

Inspect the claim, method and frozen source material.

Historical naming note. The published study used earlier internal product names and predates Frame. It is evidence about the tested method and condition, not a test of the current Frame Free plan or the future Frame Pro plan.

Future work needs to widen the test: more decision domains, more users, sharper measurement of the outcome.

Future studies should test different decision domains, model environments and user populations, with clearer measurement of downstream action and decision quality.