PAN simulation results4

In the sociotechnical simulation, over a supervised-plus-agent scenario, adding a verifier to the autonomous agent removed roughly 46% of the harm that persists and a coordinated governance package roughly 43%, while upgrading the model alone removed only about 6%.

Sociotechnical simulation result: PAN social-work governance guidance, lever-ranking comparison.

Appears on: /practice/verifier-on-the-agent, /practice/improve-the-model, /pan-lab

In the sociotechnical simulation, fixing the surrounding system out-leveraged an equal-effort model upgrade in nearly every case tested, and by several times the margin - a better model helps least where the system, not the model, does the damage.

Sociotechnical simulation result: PAN baseline analysis.

Appears on: /practice/improve-the-model, /pan-lab

In the sociotechnical simulation, deleting records without reading them raised the contaminated share by stripping out benign entries; only content-aware cleanup reliably reduced it.

Sociotechnical simulation result: PAN governance-lever audit.

Appears on: /practice/connection-authorization, /pan-lab, /practice/data-minimization, /practice/content-aware-decontamination, /practice/record-reconciler

In the sociotechnical simulation, the same AI in three modeled office cultures - stylized, not real workplaces - let errors stick at very different rates: roughly 75% under low-oversight autonomy, 20% under human supervision, and 16% under high-governance professional controls.

Sociotechnical simulation result: PAN social-work governance guidance, three-office comparison.

Appears on: /pan-lab