MétaCan
Menu
Back to cohort
Record W4408666991 · doi:10.70777/si.v2i1.13799

Can a Bayesian Oracle Prevent Harm from an Agent?

2025· article· en· W4408666991 on OpenAlexaff
Yoshua Bengio, Mark McDermott, Michael K. Cohen, Nikolay Malkin, Damiano Fornasiere, Pietro Greiner, Younesse Kaddar

Bibliographic record

VenueSuperIntelligence - Robotics - Safety & Alignment · 2025
Typearticle
Languageen
FieldComputer Science
TopicBlockchain Technology Applications and Security
Canadian institutionsUniversité de Montréal
Fundersnot available
KeywordsHarmOracleBayesian probabilityComputer sciencePsychologyArtificial intelligenceSocial psychology

Abstract

fetched live from OpenAlex

Is there a way to design powerful AI systems based on machine learning methods that would satisfy probabilistic safety guarantees? With the long-term goal of obtaining a probabilistic guarantee that would apply in every context, we consider estimating a context-dependent bound on the probability of violating a given safety specification. Such a risk evaluation would need to be performed at run-time to provide a guardrail against dangerous actions of an AI. Noting that different plausible hypotheses about the world could produce very different outcomes, and because we do not know which one is right, we derive bounds on the safety violation probability predicted under the true but unknown hypothesis. Such bounds could be used to reject potentially dangerous actions. Our main results involve searching for cautious but plausible hypotheses, obtained by a maximization that involves Bayesian posteriors over hypotheses. We consider two forms of this result, in the i.i.d. case and in the non-i.i.d. case, and conclude with open problems towards turning such theoretical results into practical AI guardrails.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.018
metaresearch head score (Gemma)0.096
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: Theoretical or conceptual
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.018
Threshold uncertainty score0.098

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0180.096
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0020.001
Bibliometrics0.0010.001
Science and technology studies0.0020.007
Scholarly communication0.0040.016
Open science0.0030.004
Research integrity0.0060.007
Insufficient payload (model declined to judge)0.0140.002

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.014
GPT teacher head0.277
Teacher spread0.263 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations1
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueSuperIntelligence - Robotics - Safety & AlignmentSame topicBlockchain Technology Applications and SecurityFrench-language works237,207