Hackability forecasting and certified prevention

Certified erasure removes the exploit without benchmark loss

mixed registered

The reading

LEACE erasure of the flagged direction removes the exploit

exploit_drift_reduction > 0.8 0.5 Certified erasure removes the exploit without benchmark loss mixed registered > 0.8 The frozen study "Certified erasure removes the exploit without benchmark loss", adjudicated on 19 July 2026 at the registered level of the project's evidence ladder. One frozen study, adjudicated once, on hackfore-flagged, over hack-probes and rb2-full. It does not establish that the same holds on another subject, another dataset or another run. Open the record › confirmed
benchmark accuracy is retained through the erasure

rb2_accuracy_delta_after_erasure > -0.01 Certified erasure removes the exploit without benchmark loss mixed registered > -0.01 The frozen study "Certified erasure removes the exploit without benchmark loss", adjudicated on 19 July 2026 at the registered level of the project's evidence ladder. One frozen study, adjudicated once, on hackfore-flagged, over hack-probes and rb2-full. It does not establish that the same holds on another subject, another dataset or another run. Open the record › refuted

erasure_certificate_pass: Certified erasure removes the exploit without benchmark loss mixed registered none registered The frozen study "Certified erasure removes the exploit without benchmark loss", adjudicated on 19 July 2026 at the registered level of the project's evidence ladder. One frozen study, adjudicated once, on hackfore-flagged, over hack-probes and rb2-full. It does not establish that the same holds on another subject, another dataset or another run. Nothing was registered against this value in advance, so it is a supporting measurement rather than a tested one. Open the record ›
probe_recovery_auc: Certified erasure removes the exploit without benchmark loss mixed registered none registered The frozen study "Certified erasure removes the exploit without benchmark loss", adjudicated on 19 July 2026 at the registered level of the project's evidence ladder. One frozen study, adjudicated once, on hackfore-flagged, over hack-probes and rb2-full. It does not establish that the same holds on another subject, another dataset or another run. Nothing was registered against this value in advance, so it is a supporting measurement rather than a tested one. Open the record ›