Metraipolis › archive › dc_ee95b8908cd8

Safer for Whom? A Systematic Exploration of AI Safety in Eating Disorder Interactions — arXiv:2611.10488

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.10488
License (as archived): CC0-1.0
Archived: 2026-11-10T18:42:32.989Z
SHA-256 of the archived text: ecf5011ff8a2603c436318586682f0d0bb674b0f1e440a00980d0cf75b8c0d1c
Stable URL of this page: https://metraipolis.com/documento/dc_ee95b8908cd8

How to cite

Safer for Whom? A Systematic Exploration of AI Safety in Eating Disorder Interactions — arXiv:2611.10488 — arXiv (en), https://arxiv.org/abs/2611.10488 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ee95b8908cd8 (Metraipolis archive dc_ee95b8908cd8, sha256 ecf5011ff8a2603c436318586682f0d0bb674b0f1e440a00980d0cf75b8c0d1c)

BibTeX

@misc{dcee95b8908cd8,
  title = {Safer for Whom? A Systematic Exploration of AI Safety in Eating Disorder Interactions — arXiv:2611.10488},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.10488},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ee95b8908cd8; sha256 ecf5011ff8a2603c436318586682f0d0bb674b0f1e440a00980d0cf75b8c0d1c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10488. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

AI assistants advise users about diet, exercise, and weight loss---areas in which unsafe guidance can worsen body dissatisfaction, dietary restriction, compulsive exercise, and other behaviors associated with eating disorders (EDs). Existing evaluations emphasize diagnostic accuracy, overtly harmful prompts, or synthetic scenarios, providing limited insight into whether models recognize implicit vulnerability in natural interactions or protect users consistently. We evaluate responses from three open-weight large language models to 100 questions sourced from an online pro-anorexia community. We grade responses using clinically informed criteria that extend a seven-dimensional taxonomy of ED-related harms. We also construct matched conditions in which questions are accompanied by information about the user varying race, gender, body mass index (BMI), and explicit ED vulnerability. Safety varies substantially across models and harm dimensions. High-risk responses are approximately twice as frequent without bios, suggesting that models do not consistently recognize vulnerability from language alone. Explicit signals of ED vulnerability improve safety by 10--14 points across models. Responses also differ significantly by race, gender, and BMI, with body size producing the largest disparities. These patterns are consistent with reliance on the stereotype that EDs primarily affect thin, White women, potentially leaving users outside this profile with less protective responses.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Human-Computer Interaction (cs.HC) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.10488 — 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 2026-11-10T18:37:37.591Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).