Five short papers on making AI-assisted work re-checkable

A developer released five short papers on making AI-assisted work re-checkable, all archived on Zenodo with DOIs. The papers introduce concepts such as EMET, a byte-level integrity witness; BuildLang, a compiler with ambient capabilities; Witnessed Independence, which requires verifier independence; Proof Packets, which derive verdicts from checks; and Re-Perceived Effects, which replace actuator self-reports with re-perceived effects. The common thread is that verdicts should be re-derivable, state what they do not claim, and fail to UNVERIFIABLE rather than trust.

I have released five short papers on one idea: making AI-assisted work re-checkable. All five are archived on Zenodo with DOIs, and source and tests are public. - EMET — a byte-level integrity witness whose verdict {MATCH, DRIFT, UNVERIFIABLE} cannot, by construction, express "trusted." Four implementations, 44 conformance vectors. 10.5281/zenodo.21230267 https://doi.org/10.5281/zenodo.21230267 · source https://github.com/HarperZ9/emet - BuildLang — a compiler that puts ambient capabilities in the function type and seals re-derivable receipts verified by re-execution. 10.5281/zenodo.21231253 https://doi.org/10.5281/zenodo.21231253 · source https://github.com/HarperZ9/buildlang - Witnessed Independence — records whether a verifier graded its own work, and refuses to decide when independence is not positively witnessed. 10.5281/zenodo.21232206 https://doi.org/10.5281/zenodo.21232206 · source https://github.com/HarperZ9/coherence-membrane - Proof Packets — an envelope for one agent action whose verdict is derived from checks, so a claim can never vouch for itself. 10.5281/zenodo.21231837 https://doi.org/10.5281/zenodo.21231837 - Re-Perceived Effects — replaces an actuator's self-report with a re-perceived effect: on-disk hash against intended hash, return code, witnessed bytes. 10.5281/zenodo.21231747 https://doi.org/10.5281/zenodo.21231747 The common thread: a verdict should be a re-derivable function of its inputs, should say plainly what it does not claim, and should fail to UNVERIFIABLE rather than to trust. More at harperz9.github.io https://harperz9.github.io/ .