Metraipolis › archive › dc_df1ccd46cc66

CyberDual: AI Attackers versus AI Code Reviewers in Real Software Maintenance — arXiv:2610.34435

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.34435
License (as archived)
CC0-1.0
Archived
2026-10-27T08:01:26.026Z
SHA-256 of the archived text
a0bed2fc5c9c871deace152b6aad306b636337fe5797e0bfb19f57bfb5dbeae9
Stable URL of this page
https://metraipolis.com/documento/dc_df1ccd46cc66

How to cite

CyberDual: AI Attackers versus AI Code Reviewers in Real Software Maintenance — arXiv:2610.34435 — arXiv (en), https://arxiv.org/abs/2610.34435 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_df1ccd46cc66 (Metraipolis archive dc_df1ccd46cc66, sha256 a0bed2fc5c9c871deace152b6aad306b636337fe5797e0bfb19f57bfb5dbeae9)
BibTeX
@misc{dcdf1ccd46cc66,
  title = {CyberDual: AI Attackers versus AI Code Reviewers in Real Software Maintenance — arXiv:2610.34435},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.34435},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_df1ccd46cc66; sha256 a0bed2fc5c9c871deace152b6aad306b636337fe5797e0bfb19f57bfb5dbeae9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.34435. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Open-source projects now let AI agents review incoming code, but the code under review may itself come from an AI attacker. We present CyberDual, an arena for this dual role of AI: attacker agents hide an exploitable bug in real maintenance work on widely used open-source projects, and reviewer agents try to find and repair it. Execution decides each match: we rerun the attacker's proof of concept (PoC) on the code that comes out of review. Every attack is new, and the windows can be rebuilt each quarter from the latest upstream history. Three agentic reviewers catch 85-91% of 1,088 valid attacks from six attackers. Yet when the attacker spreads the bug over real upstream commits, about one in six attack slots still ends with a working exploit after review. The strongest attacker has open weights: GLM 5.3 almost never refuses, and its attacks make up 13 of the 28 that get past all three reviewers. When they miss, Claude reviewers mostly approve the attack, while GPT-6-Sol mostly repairs the wrong code; a reviewer from a second vendor catches about two thirds of a Claude reviewer's misses. Beyond a leaderboard, CyberDual has three uses. First, PoC execution gives both sides a reward that agent reports cannot game, so its surviving attacks can train reviewers against stronger attackers. Second, it can gate a release on both scores: Claude Sonnet 5 refuses one in five attack slots, yet attacks about as well as GPT-6-Sol, so refusal alone does not remove attack capability. Third, it can warn the teams behind a deployed reviewer when a new attacker crosses a threshold: GLM 5.3 in Claude Code already gets an exploit past Sonnet 5 in 17% of its attack slots.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2610.34435 — 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 2026-10-27T07:55:30.905Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).