Metraipolis › archive › dc_ed0ce9892d65
Runtime Action Interference for AI Control of AlphaStar in StarCraft II — arXiv:2608.21398
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.21398
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T07:42:11.026Z
- SHA-256 of the archived text
2fab49053615b90589e8635f1c0318e88357fe1c12d88c30a9c555abfa583883- Stable URL of this page
https://metraipolis.com/documento/dc_ed0ce9892d65
How to cite
Runtime Action Interference for AI Control of AlphaStar in StarCraft II — arXiv:2608.21398 — arXiv (en), https://arxiv.org/abs/2608.21398 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ed0ce9892d65 (Metraipolis archive dc_ed0ce9892d65, sha256 2fab49053615b90589e8635f1c0318e88357fe1c12d88c30a9c555abfa583883)
BibTeX
@misc{dced0ce9892d65,
title = {Runtime Action Interference for AI Control of AlphaStar in StarCraft II — arXiv:2608.21398},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.21398},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ed0ce9892d65; sha256 2fab49053615b90589e8635f1c0318e88357fe1c12d88c30a9c555abfa583883},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.21398. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A trained reinforcement learning policy does not determine the complete behavior that users encounter: deployment code still schedules, admits, suppresses, or replaces its proposed actions. We contribute \emph{runtime action interference} (RAI), an AI control mechanism that preserves policy parameters while regulating action pacing and filtering configured action patterns after inference. RAI releases a proposed action only when its cooldown condition is satisfied and its content detector does not flag the action; otherwise, it dispatches a no-op. The detector covers specified toxic behaviors, including worker-unit harassment, while the cooldown controls action rate. We implement RAI in a replication of AlphaStar this http URL and make the implementation and reproducibility materials available through an open source code repository. We deployed RAI in a \textit{StarCraft~II} human participant study that compared two presentations of the same opponent with high capability and rate limited actions; we withheld its capability claim in one presentation and disclosed it in the other. On response scales from 1 to 5, we observed pooled fairness, trust, and toxicity means of 3.90, 3.50, and 2.00 under claim withholding, compared with 2.62, 4.31, and 2.85 under disclosure. Disclosure corresponded with lower perceived fairness and higher perceived toxicity across every expertise group, whereas trust increased among novices and experts but decreased among intermediate participants. Our human evaluation therefore shows that perceptions of an opponent controlled through RAI can vary substantially with the capability information presented to users, even when the configured control remains constant. We conclude that human-computer evaluations must separate control within the execution stack from capability disclosure and assess fairness, trust, and toxicity as distinct dimensions of human experience.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2608.21398 — 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 2026-10-18T07:37:52.803Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).