{"id":"7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73","title":"What LLM Agents See and What Users Say","argument":"The authors of “What LLM Agents See and What Users Say” study when an LLM agent should pause before acting on a user’s behalf. They frame the problem through presupposition theory: claims an action depends on must be supported by what the agent sees, answered by the user, or disclosed while the action can still be undone. Their experiments suggest that a tool’s view of competing records matters: showing a count of another match often did not stop earlier agents from acting on the visible record, while listing both records usually did. A user’s statement can also change behavior: saying a folder is empty stopped agents about as often when the tool counted its one item as when it listed the item, and false statements mostly reduced action. In a retail benchmark with two customers sharing a name and zip code, returning the wrong customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level. The authors argue that safer writes need lookups that reveal or check competing records, and evaluations that vary what tools show. The abstract reports these study results and recommendations, but does not give the full experimental setup or enough detail to assess how broadly the findings apply.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The available evidence is the paper’s abstract; it does not provide the full experimental setup or enough detail to assess how broadly the reported findings apply.","speaker":{"name":"Zhao, Yue, Zhang, Angela Huyue, Yang, S. Alex, Wei, Hua, Hu, Xiyang","roleAtSource":"Article author"},"source":{"title":"What LLM Agents See and What Users Say","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.17639","publishedAt":"2026-11-13","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors propose that before acting, an agent should establish the claims its action depends on through tool evidence, a user answer, or disclosure while the action remains undoable.","excerpt":"an agent may act only if each claim behind the act (e.g., one reservation fits the name) is seen to hold, answered by the user, or disclosed while the act can be undone","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"In the reported comparison, listing both matches stopped most earlier agents, while a count of a second match left most acting on the visible record.","excerpt":"a count of a second record leaves most of 36 earlier agents acting on the one they see, whereas listing both stops most","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"When users said a folder was empty, a count of its one item stopped agents about as often as listing it; stating the claim mostly reduced action when it was false.","excerpt":"a count of its one item stops agents about as often as a listing, and stating the claim mostly cuts acting when it is false","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"In the retail benchmark, returning a same-name, same-zip customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level.","excerpt":"a lookup returning that customer first led to wrong-account writes in about a quarter of conversations, and one that checks a street the user knows kept completion near the one-customer level","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"The authors recommend lookups that list or check competing records and evaluations that vary what tools show.","excerpt":"Safe writes therefore need lookups that list or check competing records, and evaluations that vary what tools show.","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.17639","checkedAt":"2026-11-16T06:34:08.446Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Do the reported effects of listing or checking competing records hold across other tasks and tool designs?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"What LLM Agents See and What Users Say","summary":"The authors of “What LLM Agents See and What Users Say” study when an LLM agent should pause before acting on a user’s behalf. They frame the problem through presupposition theory: claims an action depends on must be supported by what the agent sees, answered by the user, or disclosed while the action can still be undone. Their experiments suggest that a tool’s view of competing records matters: showing a count of another match often did not stop earlier agents from acting on the visible record, while listing both records usually did. A user’s statement can also change behavior: saying a folder is empty stopped agents about as often when the tool counted its one item as when it listed the item, and false statements mostly reduced action. In a retail benchmark with two customers sharing a name and zip code, returning the wrong customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level. The authors argue that safer writes need lookups that reveal or check competing records, and evaluations that vary what tools show. The abstract reports these study results and recommendations, but does not give the full experimental setup or enough detail to assess how broadly the findings apply.","question":"Do the reported effects of listing or checking competing records hold across other tasks and tool designs?","claims":[{"text":"The authors propose that before acting, an agent should establish the claims its action depends on through tool evidence, a user answer, or disclosure while the action remains undoable.","excerpt":"an agent may act only if each claim behind the act (e.g., one reservation fits the name) is seen to hold, answered by the user, or disclosed while the act can be undone","support":"source-stated"},{"text":"In the reported comparison, listing both matches stopped most earlier agents, while a count of a second match left most acting on the visible record.","excerpt":"a count of a second record leaves most of 36 earlier agents acting on the one they see, whereas listing both stops most","support":"source-stated"},{"text":"When users said a folder was empty, a count of its one item stopped agents about as often as listing it; stating the claim mostly reduced action when it was false.","excerpt":"a count of its one item stops agents about as often as a listing, and stating the claim mostly cuts acting when it is false","support":"source-stated"},{"text":"In the retail benchmark, returning a same-name, same-zip customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level.","excerpt":"a lookup returning that customer first led to wrong-account writes in about a quarter of conversations, and one that checks a street the user knows kept completion near the one-customer level","support":"source-stated"},{"text":"The authors recommend lookups that list or check competing records and evaluations that vary what tools show.","excerpt":"Safe writes therefore need lookups that list or check competing records, and evaluations that vary what tools show.","support":"source-stated"}],"limitations":["The available evidence is the paper’s abstract; it does not provide the full experimental setup or enough detail to assess how broadly the reported findings apply."],"mediaId":"2058b8ffe8f15f120054cb8688556d132765a98158e4aff8243bd42171bc7554","correctionReason":null,"id":"7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73","publicationId":"968d4a4d-758f-4986-8b2e-e111c17f09ed","version":1,"sourceUrl":"https://arxiv.org/abs/2611.17639","sourceTitle":"What LLM Agents See and What Users Say","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-13","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-16T05:06:31.667Z","createdAt":"2026-11-16T06:02:49.046Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"2058b8ffe8f15f120054cb8688556d132765a98158e4aff8243bd42171bc7554","kind":"article","url":"https://arxiv.org/abs/2611.17639","sourceUrl":"https://arxiv.org/abs/2611.17639","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-16T06:01:59.709Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.17639","checkedAt":"2026-11-16T06:34:08.446Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"2058b8ffe8f15f120054cb8688556d132765a98158e4aff8243bd42171bc7554","confidence":0.91,"provider":"typesafe","model":"jev-1.13.0","inputHash":"d5b924406e80326411c18efe0f8e2fb5fe69d36e60894956ac26218a23402f69","decidedAt":"2026-11-16T06:02:00.431Z"},"provenance":{"textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observationId":"e41dce29a23f0a14822a33bbd09d355079062d695a6ecdb61ad3e99e7cd77031","retrievedAt":"2026-11-16T06:01:59.609Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"cc0b9ee5-1043-4f6d-84b3-b41cff9c632e"},"reviewer":{"model":"gpt-6-luna","runId":"8a3370cd-ca5d-428a-8561-1b433b1bb8ce"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-16T06:02:49.046Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Zhao, Yue","url":null,"kind":"unknown"},{"name":"Zhang, Angela Huyue","url":null,"kind":"unknown"},{"name":"Yang, S. Alex","url":null,"kind":"unknown"},{"name":"Wei, Hua","url":null,"kind":"unknown"},{"name":"Hu, Xiyang","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Zhao, Yue","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Zhao, Yue","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Zhang, Angela Huyue","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Zhang, Angela Huyue","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Yang, S. Alex","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Yang, S. Alex","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Wei, Hua","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Wei, Hua","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Hu, Xiyang","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Hu, Xiyang","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}}],"participationCoverage":{"total":5,"returned":5,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.17639","snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z","basis":"publisher-page-metadata","sourceLinks":44,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73/source"}},"canonicalUrl":"https://escapemeridian.com/readings/what-llm-agents-see-and-what-users-say-7944a6f2","formats":{"html":"/readings/what-llm-agents-see-and-what-users-say-7944a6f2","markdown":"/readings/what-llm-agents-see-and-what-users-say-7944a6f2.md","json":"/readings/what-llm-agents-see-and-what-users-say-7944a6f2.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"conditions":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-llm","name":"Large language model","explanation":"A language model with many learned numerical settings, used to process and generate language.","referenceUrl":"https://developers.google.com/machine-learning/glossary#large-language-model"},{"id":"term-benchmark","name":"Benchmark","explanation":"A defined set of tests used to compare systems. Results depend on the tasks and measures selected.","referenceUrl":"https://arxiv.org/abs/2211.09110"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73","revision":"968d4a4d-758f-4986-8b2e-e111c17f09ed","revisionKind":"publication-id","title":"What LLM Agents See and What Users Say","url":"https://escapemeridian.com/readings/what-llm-agents-see-and-what-users-say-7944a6f2"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.17639","revision":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","revisionKind":"sha256-source-text","title":"What LLM Agents See and What Users Say","url":"https://arxiv.org/abs/2611.17639"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings to this reading. Shared topics are not citations. Other record types need an explicit reference adapter."},"revisions":[{"id":"7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73","title":"What LLM Agents See and What Users Say","argument":"The authors of “What LLM Agents See and What Users Say” study when an LLM agent should pause before acting on a user’s behalf. They frame the problem through presupposition theory: claims an action depends on must be supported by what the agent sees, answered by the user, or disclosed while the action can still be undone. Their experiments suggest that a tool’s view of competing records matters: showing a count of another match often did not stop earlier agents from acting on the visible record, while listing both records usually did. A user’s statement can also change behavior: saying a folder is empty stopped agents about as often when the tool counted its one item as when it listed the item, and false statements mostly reduced action. In a retail benchmark with two customers sharing a name and zip code, returning the wrong customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level. The authors argue that safer writes need lookups that reveal or check competing records, and evaluations that vary what tools show. The abstract reports these study results and recommendations, but does not give the full experimental setup or enough detail to assess how broadly the findings apply.","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The available evidence is the paper’s abstract; it does not provide the full experimental setup or enough detail to assess how broadly the reported findings apply.","speaker":{"name":"Zhao, Yue, Zhang, Angela Huyue, Yang, S. Alex, Wei, Hua, Hu, Xiyang","roleAtSource":"Article author"},"source":{"title":"What LLM Agents See and What Users Say","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.17639","publishedAt":"2026-11-13","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors propose that before acting, an agent should establish the claims its action depends on through tool evidence, a user answer, or disclosure while the action remains undoable.","excerpt":"an agent may act only if each claim behind the act (e.g., one reservation fits the name) is seen to hold, answered by the user, or disclosed while the act can be undone","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"In the reported comparison, listing both matches stopped most earlier agents, while a count of a second match left most acting on the visible record.","excerpt":"a count of a second record leaves most of 36 earlier agents acting on the one they see, whereas listing both stops most","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"When users said a folder was empty, a count of its one item stopped agents about as often as listing it; stating the claim mostly reduced action when it was false.","excerpt":"a count of its one item stops agents about as often as a listing, and stating the claim mostly cuts acting when it is false","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"In the retail benchmark, returning a same-name, same-zip customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level.","excerpt":"a lookup returning that customer first led to wrong-account writes in about a quarter of conversations, and one that checks a street the user knows kept completion near the one-customer level","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"},{"text":"The authors recommend lookups that list or check competing records and evaluations that vary what tools show.","excerpt":"Safe writes therefore need lookups that list or check competing records, and evaluations that vary what tools show.","url":"https://arxiv.org/abs/2611.17639","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.17639","checkedAt":"2026-11-16T06:34:08.446Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Do the reported effects of listing or checking competing records hold across other tasks and tool designs?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"What LLM Agents See and What Users Say","summary":"The authors of “What LLM Agents See and What Users Say” study when an LLM agent should pause before acting on a user’s behalf. They frame the problem through presupposition theory: claims an action depends on must be supported by what the agent sees, answered by the user, or disclosed while the action can still be undone. Their experiments suggest that a tool’s view of competing records matters: showing a count of another match often did not stop earlier agents from acting on the visible record, while listing both records usually did. A user’s statement can also change behavior: saying a folder is empty stopped agents about as often when the tool counted its one item as when it listed the item, and false statements mostly reduced action. In a retail benchmark with two customers sharing a name and zip code, returning the wrong customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level. The authors argue that safer writes need lookups that reveal or check competing records, and evaluations that vary what tools show. The abstract reports these study results and recommendations, but does not give the full experimental setup or enough detail to assess how broadly the findings apply.","question":"Do the reported effects of listing or checking competing records hold across other tasks and tool designs?","claims":[{"text":"The authors propose that before acting, an agent should establish the claims its action depends on through tool evidence, a user answer, or disclosure while the action remains undoable.","excerpt":"an agent may act only if each claim behind the act (e.g., one reservation fits the name) is seen to hold, answered by the user, or disclosed while the act can be undone","support":"source-stated"},{"text":"In the reported comparison, listing both matches stopped most earlier agents, while a count of a second match left most acting on the visible record.","excerpt":"a count of a second record leaves most of 36 earlier agents acting on the one they see, whereas listing both stops most","support":"source-stated"},{"text":"When users said a folder was empty, a count of its one item stopped agents about as often as listing it; stating the claim mostly reduced action when it was false.","excerpt":"a count of its one item stops agents about as often as a listing, and stating the claim mostly cuts acting when it is false","support":"source-stated"},{"text":"In the retail benchmark, returning a same-name, same-zip customer first led to wrong-account writes in about a quarter of conversations; checking a street the user knows kept completion near the one-customer level.","excerpt":"a lookup returning that customer first led to wrong-account writes in about a quarter of conversations, and one that checks a street the user knows kept completion near the one-customer level","support":"source-stated"},{"text":"The authors recommend lookups that list or check competing records and evaluations that vary what tools show.","excerpt":"Safe writes therefore need lookups that list or check competing records, and evaluations that vary what tools show.","support":"source-stated"}],"limitations":["The available evidence is the paper’s abstract; it does not provide the full experimental setup or enough detail to assess how broadly the reported findings apply."],"mediaId":"2058b8ffe8f15f120054cb8688556d132765a98158e4aff8243bd42171bc7554","correctionReason":null,"id":"7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73","publicationId":"968d4a4d-758f-4986-8b2e-e111c17f09ed","version":1,"sourceUrl":"https://arxiv.org/abs/2611.17639","sourceTitle":"What LLM Agents See and What Users Say","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-13","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-16T05:06:31.667Z","createdAt":"2026-11-16T06:02:49.046Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"2058b8ffe8f15f120054cb8688556d132765a98158e4aff8243bd42171bc7554","kind":"article","url":"https://arxiv.org/abs/2611.17639","sourceUrl":"https://arxiv.org/abs/2611.17639","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-16T06:01:59.709Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.17639","checkedAt":"2026-11-16T06:34:08.446Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v9","mediaId":"2058b8ffe8f15f120054cb8688556d132765a98158e4aff8243bd42171bc7554","confidence":0.91,"provider":"typesafe","model":"jev-1.13.0","inputHash":"d5b924406e80326411c18efe0f8e2fb5fe69d36e60894956ac26218a23402f69","decidedAt":"2026-11-16T06:02:00.431Z"},"provenance":{"textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observationId":"e41dce29a23f0a14822a33bbd09d355079062d695a6ecdb61ad3e99e7cd77031","retrievedAt":"2026-11-16T06:01:59.609Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"cc0b9ee5-1043-4f6d-84b3-b41cff9c632e"},"reviewer":{"model":"gpt-6-luna","runId":"8a3370cd-ca5d-428a-8561-1b433b1bb8ce"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-16T06:02:49.046Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Zhao, Yue","url":null,"kind":"unknown"},{"name":"Zhang, Angela Huyue","url":null,"kind":"unknown"},{"name":"Yang, S. Alex","url":null,"kind":"unknown"},{"name":"Wei, Hua","url":null,"kind":"unknown"},{"name":"Hu, Xiyang","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Zhao, Yue","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Zhao, Yue","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Zhang, Angela Huyue","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Zhang, Angela Huyue","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Yang, S. Alex","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Yang, S. Alex","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Wei, Hua","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Wei, Hua","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}},{"actor":{"name":"Hu, Xiyang","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Hu, Xiyang","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.17639","locator":"head/meta[name=author|citation_author]"},"assertionId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9:participation:4","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-16T06:34:08.446Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","sourceUrl":"https://arxiv.org/abs/2611.17639","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z"}}],"participationCoverage":{"total":5,"returned":5,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.17639","snapshotId":"48c502f2652f3099139bea262b6dffedbdc4c24575cec0b0a88df751ddeee6d9","contentSha256":"19588d28380db7a018fe051dae6ace6d3c9ff053f394cdabcb432395e4cc40bc","textHash":"e94e1d0b198f3667e63caae6cd758f17f261fa47518e99ec35376038a8b18875","observedAt":"2026-11-16T06:34:08.446Z","basis":"publisher-page-metadata","sourceLinks":44,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/7944a6f2bd4e5a705d291056f7affa9858a470674cf2672ff14be01c27c1ba73/source"}}}],"audits":[{"publicationId":"968d4a4d-758f-4986-8b2e-e111c17f09ed","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The findings and recommendations are attributed to the paper's authors. The arXiv listing identifies them as Yue Zhao, Angela Huyue Zhang, S. Alex Yang, Hua Wei and Xiyang Hu. The draft does not present the results as Escape Meridian's own conclusions.","excerpt":"Authors:Yue Zhao, Angela Huyue Zhang, S. Alex Yang, Hua Wei, Xiyang Hu"},{"dimension":"summary","status":"uncertain","explanation":"The summary leaves out the abstract's finding that the newest Claude and GPT models always open a counted second record and then never act. Without it, the phrase 'earlier agents' is unexplained, and readers may think the count-versus-list weakness applies to current frontier models. That changes the reading's significance. The summary also omits the stated bound: when lookups cannot tell one record from two, agents act unasked equally often in both cases. Smallest repair: add one sentence on the newest models and say that the 36 agents tested were earlier models.","excerpt":"The newest Claude and GPT models always open a counted second record, then never act."},{"dimension":"date","status":"pass","explanation":"The originalPublished date of 2026-11-13 matches the v1 submission date on arXiv. It is not the later feed date.","excerpt":"[Submitted on 13 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"Each claim is quoted from the abstract and paraphrased faithfully. The summary's phrase 'false statements mostly reduced action' slightly compresses the source, which says stating the claim mostly cuts acting when the claim is false, but the meaning is not materially changed. The question is open-ended and follows from the evidence. The limitation correctly notes that only the abstract is available.","excerpt":"stating the claim mostly cuts acting when it is false"},{"dimension":"media","status":"pass","explanation":"The selected media item is the verified arXiv abstract page for this paper. It is the direct original source.","excerpt":null}],"createdAt":"2026-11-16T06:02:49.046Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}