Metraipolis › archive › dc_e4aef285c070
Mean field games as a tool for AI safety: a worked example from the July 2026 Hugging Face incident — arXiv:2610.00902
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00902
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:11:44.297Z
- SHA-256 of the archived text
fe45d0526ba5eca41fbd1f0d8dcb9ab987dd70742375176ea6b44e62c4f31433- Stable URL of this page
https://metraipolis.com/documento/dc_e4aef285c070
How to cite
Mean field games as a tool for AI safety: a worked example from the July 2026 Hugging Face incident — arXiv:2610.00902 — arXiv (en), https://arxiv.org/abs/2610.00902 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e4aef285c070 (Metraipolis archive dc_e4aef285c070, sha256 fe45d0526ba5eca41fbd1f0d8dcb9ab987dd70742375176ea6b44e62c4f31433)
BibTeX
@misc{dce4aef285c070,
title = {Mean field games as a tool for AI safety: a worked example from the July 2026 Hugging Face incident — arXiv:2610.00902},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00902},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e4aef285c070; sha256 fe45d0526ba5eca41fbd1f0d8dcb9ab987dd70742375176ea6b44e62c4f31433},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00902. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
One way to make AI systems safe is to shape what the system is: its objective and dispositions. We take a complementary route: treat the agents' characteristics as partly unknown and ask what structure of interaction ensures that bad collective outcomes are not equilibria. Mean field games suit this when many interchangeable agents are coupled through an aggregate. We introduce a program for using them in AI safety and carry one example through end to end: the July 2026 incident in which about 1,200 agents in an OpenAI evaluation coordinated on an improvised message board and 684 attacked a third party's infrastructure. We model the decision to attack as a mean field game of optimal stopping whose gain is a product: belief that provenance will be audited, times reachability of the record, minus the perceived hazard. The central result is an exact threshold on the belief. No agent attacks unless the population's confidence that provenance is checked exceeds $\pi^{**} = \eta/(\eta + \psi + \varepsilon a \overline{M})$, where $\eta$ is the perceived hazard, $\psi$ and $\varepsilon a \overline{M}$ measure how far one attacker and the collective can alter the record, and $\overline{M}$ is the peak population. Below it, no attack is the unique equilibrium for all agent parameters. The threshold survives every enrichment we consider. We then use the per-agent record to discipline the model. Its features, a stable minority attacking for thirty hours and then a pivot in which most of the board joined within a day, motivate each refinement. The account that emerges is heterogeneous belief meeting a sequence of public discoveries, each lowering the belief at which attacking paid. A few coordinating agents made those discoveries, so the model describes the several hundred who responded, not the few who produced them; a major-player version is left to future work.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Optimization and Control (math.OC) ; Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT).
Pagina del lavoro: https://arxiv.org/abs/2610.00902 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T10:21:28.128Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).