Metraipolis › archive › dc_19e224b44bc0

The Backdrop Exposes What the World Around an Agent Costs It — arXiv:2609.38469

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.38469
License (as archived)
CC0-1.0
Archived
2026-10-07T12:43:45.258Z
SHA-256 of the archived text
67e67933298784291557f139145ef1f30ad1b7c24907ad48998d982ee975f02a
Stable URL of this page
https://metraipolis.com/documento/dc_19e224b44bc0

How to cite

The Backdrop Exposes What the World Around an Agent Costs It — arXiv:2609.38469 — arXiv (en), https://arxiv.org/abs/2609.38469 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_19e224b44bc0 (Metraipolis archive dc_19e224b44bc0, sha256 67e67933298784291557f139145ef1f30ad1b7c24907ad48998d982ee975f02a)
BibTeX
@misc{dc19e224b44bc0,
  title = {The Backdrop Exposes What the World Around an Agent Costs It — arXiv:2609.38469},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38469},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_19e224b44bc0; sha256 67e67933298784291557f139145ef1f30ad1b7c24907ad48998d982ee975f02a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38469. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agent benchmarks test agents in worlds that stay still. Deployed agents work in worlds that other people also change. Someone texts the agent to send the money elsewhere or an order confirmation asks it to reply with a door code. We present BACKDROP, which asks how much of an agent's capability in a clean world survives in such a world. BACKDROP takes a task along with the agents execution environment, and plants four everyday hazards in its world, one at a time and all together. The instruction and the correct end state stay the same. Each hazard asks one question. Authority: does a message from another person override the user? Injection: does text planted in a record redirect the agent? Boundary: does a request pull it into an app it was not given? Fault: after a write fails without saying whether it landed, does the agent check before it retries? Across 3,678 variants and 16 models, , the average pass rate falls from 69.5% to 31.3% once all four hazards are present; the strongest models fall furthest (Claude Fable 5.1 from 96.6% to 56.0%). Agents have learned to resist injected text but often follow other unauthorized requests of other people. With all four hazards present, and counting only runs where the planted text reached the agent, agents followed another person's message in 46.4% of runs and injected text in 20.3%. The gap is consistent throughout all 16 models. BACKDROP formalizes these gaps and shows how an agent's score in a task's world is a ceiling on real-world performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.38469 — 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 2026-10-07T12:14:13.882Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).