P094 Purpose-level (semantic) use verification

Can any on-chip-measurable feature verify the purpose or beneficiary of a workload, not just its type?

Verifying the purpose or beneficiary of a workload, not just its type.

Status

Posed

Posed by

Also raised in: Reuel, Bucknall et al. (2025), Open Problems in Technical AI Governance, compute sections; Harack et al. (2025), Verification for International AI Governance (Oxford AIGI); Aarne & Petrie (2025), International Security Applications of Flexible Hardware-Enabled Guarantees (flexHEG Part 3); Scher (2026), Verifying Restrictions on Frontier AI Research (MIRI); Cankaya (2026), A System Overview for Near-Term, Low-Trust AI Compute Verification (v0.2).

Sub-questions

IDSub-questionWhat the sources report
P094.1Has a public / shared compute programme verified declared purpose or prevented resale of access without exposing user code or data?open

Sub-questions not yet stated precisely enough to check

IDSub-question
P094.2For a named pair of uses that differ only in purpose (e.g. patching vs exploitation, drone control), does any on-chip-measurable feature separate them above chance?
P094.3Can any mechanism verify the de facto beneficiary of a workload?

Possibly relevant work found by automated search

Found by AI agents searching for this problem's sub-questions, mostly by keyword and citation in the OpenAlex scholarly database and on the web; a second AI pass reviewed each one, and we spot-checked some by hand. They seemed reasonably related, as far as one can tell from outside each paper's field. Some come from neighbouring fields, such as arms control and financial audit, and bear on the problem by analogy. Not reviewed against the status above; a paper listed here does not change that status. 928 of 1052 quotes checked word for word against the paper's text.

For P094.1: Has a public / shared compute programme verified declared purpose or prevented resale of access without exposing user code or data?

Data