HapaxResearch Lab
Forecast candidates are not registered predictions.

prediction / pred-2026-0003

Joint evaluation report

Draft · not registered

By 2027-09-30, NAAIMES will publish or formally co-issue a joint evaluation report identifying at least three member institutes, naming at least three model versions from at least two developers, reporting a common numeric model-level metric for every named model, and disclosing the task set, scoring rule, model/version, and principal inference or elicitation settings.

NAAIMES, as named by its source: “the International Network for Advanced AI Measurement, Evaluation and Science (NAAIMES) – formerly the International Network of AI Safety Institutes” (UK AI Security Institute, “International evaluation best practice and open questions in AI measurement”, retrieved ). This expansion is not part of the proposition.

The resolver for this forecast is not yet bound. The linked source page is the U.S. CAISI centre, not the network’s archive.

42% candidate probability. Not registered. No outcome or prospective score is claimed.

Open · source report · forecast · not verified

Evidence
source report · single-elicitor forecast
Conclusion
forecast (not a grade)
Checked
Not verified
Scope
HRL-inferred prospective forecast over public institutional records; no safety-outcome inference.

Draft candidate, not public preregistration. Full resolution cards, source snapshots, attribution, baselines, independent checker, and publication admission are outstanding.

The proposition and candidate probability are reproduced from the frozen six-entry draft slate. Inclusion here does not register the prediction. All dates and resolution details require the full registration card before prospective scoring begins.

Resolution and registration state

By 2027-09-30, NAAIMES will publish or formally co-issue a joint evaluation report identifying at least three member institutes, naming at least three model versions from at least two developers, reporting a common numeric model-level metric for every named model, and disclosing the task set, scoring rule, model/version, and principal inference or elicitation settings.

Deadline
· candidate normalization; full card review pending
Registration
Not registered
Baseline
Not supplied
Checker
Not appointed in this record

Full scoring and ambiguity protocol

Correction history

No correction to this record has been published.

Record identity

SHA-256 of the Markdown source, including its frontmatter:

37fe328fa31d1664b8c9622d89cda62d482dd087f09921fa151ec9654eefe9d4

A digest identifies bytes. It does not verify truth or certify publication.

Download the record dataset