Metraipolis › archive › dc_99c5072ff8db

arXiv 2610.30365: Data-Void Exploits in Web-Searching LLMs

Source
Metraipolis archive
Original
https://arxiv.org/abs/2610.30365
License (as archived)
CC-BY-4.0
Archived
2026-10-25T12:11:17.333Z
SHA-256 of the archived text
a4a953100ac18a4e31a3ea1c7bbc17c8521124cca62b5f48b64fb4686b2746c6
Stable URL of this page
https://metraipolis.com/documento/dc_99c5072ff8db

How to cite

arXiv 2610.30365: Data-Void Exploits in Web-Searching LLMs — Metraipolis archive, https://arxiv.org/abs/2610.30365 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_99c5072ff8db (Metraipolis archive dc_99c5072ff8db, sha256 a4a953100ac18a4e31a3ea1c7bbc17c8521124cca62b5f48b64fb4686b2746c6)
BibTeX
@misc{dc99c5072ff8db,
  title = {arXiv 2610.30365: Data-Void Exploits in Web-Searching LLMs},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.30365},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_99c5072ff8db; sha256 a4a953100ac18a4e31a3ea1c7bbc17c8521124cca62b5f48b64fb4686b2746c6},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.30365. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Web-searching Large Language Models (LLMs) have changed how users seek online information: instead of sifting through ranked lists of links, users read synthesized answers. This shift creates a new LLM vulnerability: a data-void exploit. Disinformers publish their narratives in search spaces where relevant and reliable content is scarce, i.e. data voids, then rely on the LLM's web-search to retrieve and then use, cite, or list those pages in its responses to data-void seeking prompts. Unlike classical search, where users see ranking cues and competing results, LLMs can obscure when an answer relies on sparse, unverified, or adversarial sources. We study this threat by varying the user's prompt tone - skeptical, excited, or neutral - and LLM choice - ChatGPT, Gemini, or Sonar Pro - as well as three attacker controls. The attacker can use one of five known strategies to construct and fill a data void with their narrative, host their content at varying levels of visibility from well-established, indexed domains to less established ones, and include generative-optimization engine features thought to boost citing by web-searching LLMs. We find that data-void exploits can succeed with only a single attacker controlled page. Across all conditions, attacker-content appears in LLM responses 31% of the time. User controls offer limited protection: while skeptical prompts slightly reduce success and slightly increase flagging or warnings in LLM responses, we find no evidence of large effects of prompt tone or model choice on success rates, although smaller effects cannot be ruled out. Overall, our results establish data voids as a practical inference-time attack surface for manipulating web-searching LLMs through strategically placed web content and highlight the urgent need for defenses against such exploits.

Articolo: «Data-Void Exploits in Web-Searching LLMs» — Guerra, Michele; Jobe, Wuyeh; Nambiar, Sanjana; Pöpper, Christina; Abouzied, Azza (2026-10-21), arXiv:2610.30365, categoria cs.CR (sicurezza e crittografia).
Fonte: https://arxiv.org/abs/2610.30365
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-25T12:11:12Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.