Source · Tier B · Technical reportSource · Mechanisms to Verify International Agreements About AI Development
Mechanisms to Verify International Agreements About AI Development
A. Scher, L. Thiergart. 2024. Machine Intelligence Research Institute.
| Original | https://arxiv.org/abs/2506.15867 |
|---|---|
| DOI | 10.48550/arXiv.2506.15867 |
| arXiv | 2506.15867 |
| Version | MIRI's Technical Governance Team lists the report among its technical reports, dated 27 November 2024 (checked 2026-10-17). The text linked here is arXiv v1, posted on 18 June 2025. |
| Accessed | 2026-10-08 |
| Authored or published by | Machine Intelligence Research Institute |
| Imported from | hodgkins-ai-verification-papers@c71e59ff0e8e |
| Note | Listed under "Research agendas and surveys" in the Hodgkins bibliography (CC BY 4.0). |
Cited by
- Deploy only evaluated modelsGoal
- Pause frontier AI developmentGoal
- Prevent weight theftGoal
- Proofs of useful work for capacity accountingMechanism
- Communication between compute groups is boundedClaim
- Chips are where they are declared to beClaim
- Compute stock is at most a declared amountClaim
- Declared hardware is idle or shut downClaim
- This compute runs inference, not trainingClaim
- There is no undeclared relevant computeClaim
- Model weights have not left the facilityClaim
- CompartmentalizationConcept
- Evidence bindingConcept
- Inference and training workloadsConcept
- Interconnect bandwidthConcept
- Positive and negative claimsConcept
- Proof of (useful) workConcept
- Remote attestationConcept
- Side channelConcept
- Undeclared computeConcept
- Machine Intelligence Research InstituteOrganization