{"as_of":"2026-08-11T18:25:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:8fa70bc233ad4eec3356887b4e85b9d47ade06969b5eb5dfb953ccc17fe38c8e","coverage":[{"denominator":50,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":50,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-28T01:48:56.367899Z","state":"measured"},{"denominator":51,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":51,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-11T06:34:44.6726+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-01T12:50:00.612066Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2606.05647","snapshot_observed_at":"2026-08-01T12:50:00.612066Z","title":"Coding with \"Enemy\": Can human developers detect AI agent sabotage?arXiv preprint arXiv:2606.05647, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19321","last_updated":"2026-07-29T16:23:28Z","snapshot_observed_at":"2026-08-03T00:38:29.761714Z","submitted_at":"2026-07-21T17:41:12Z","title":"ResearchArena: Evaluating Sabotage and Monitoring in Automated AI R&D","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-01T12:50:00.612066Z"},"links":{"cited_paper":"/paper/2606.05647","citing_paper":"/paper/2607.19321"},"observation_digest":"sha256:934ac26eeb9dc7663067411cd99048c78b4fa5377ebc01a55e8b1d279faf3dda","observation_id":"d684a475-b17e-4e4b-977a-6aac26fb3534","resolution":{"observed_at":"2026-08-01T12:50:00.612066Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2606.05647/citation-record","integrity":"/paper/2606.05647/integrity","json":"/paper/2606.05647/citation-record.json","paper":"/paper/2606.05647"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Information and Software Technology , pages=","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:a2a35c89733e6a95ab358f7e1736278a35cd9557d2e41f4b4edf9fefeeeefa3c","observation_id":"c5cf12ac-2454-4d40-bc2c-7f7c10b1fa0f","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"and Yang, John and Wettig, Alexander and Yao, Shunyu and Pei, Kexin and Press, Ofir and Narasimhan, Karthik , booktitle =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:7d9b70014374b0e3135d3e59fd469e46e0b34468af340741cb44bd75d5b3fec1","observation_id":"a75ae00e-f95c-4d12-bc1d-c850dd9a1104","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Advances in Neural Information Processing Systems , volume=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:8f5c55323e5ee9e4c1a52ac5e9856564d5d84c6dd8c6953bc924cf5cb58710b2","observation_id":"9cab7743-e59f-4e8a-ae1e-ee35058de401","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.18653/v1/2025.naacl-long.342","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models , booktitle =","venue":null,"work_id":"b7fd12ff-b5a5-42a1-8f83-4e0c1facc687","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:5ee488656ffd25288811b52abd6baba9496af95d68ca1ae272e16eb49da20090","observation_id":"99045116-2597-43ba-aded-dc792c07ca4e","resolution":{"observed_at":"2026-06-28T01:51:28.890593Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-07-17T19:51:00.928338+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-17T19:51:00.928338+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2512.14012","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-04T20:20:07.932018Z","title":"Retrieved from https://arxiv.org/abs/2512.14012","venue":null,"work_id":"457f9059-31e9-43cb-ac8d-ffaf6b93cbab","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:3b171bbf3b078b9fa30ad2ccbbc81ae59ec10222087766c56aa71d66b19a5920","observation_id":"36eb4624-5023-45bd-a2f4-158dd4183ff8","resolution":{"observed_at":"2026-07-02T12:46:57.264600Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2511.04824","doi":"10.48550/arxiv.2511.04824","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Horikawa, H","venue":"arXiv (Cornell University)","work_id":"fbc9fc36-39f5-4e73-8d26-8abee8d479db","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:40c680ce824bfd80a75b484c830bd389f389c4b8be5ac3d271c61f8af33123c8","observation_id":"71e53e0f-4dbb-4f5e-b36f-7a8e38b73c40","resolution":{"observed_at":"2026-07-02T12:46:57.281030Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2601.11868","last_updated":"2026-01-17T01:29:30Z","snapshot_observed_at":"2026-07-06T22:41:58.373427Z","submitted_at":"2026-01-17T01:29:30Z","title":"Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces","version":1},"cited_work":{"arxiv_id":"2601.11868","doi":"10.48550/arxiv.2302.01973","metadata_source":"pith","pith_arxiv_id":"2601.11868","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces","venue":"cs.SE","work_id":"0624be05-1d97-4fd6-8300-b04b8a3ab04b","year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2601.11868","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:2329a5fd0801a41a4477d69fd1a4154452f365df593f8d5bc82af59fe4e9f3ae","observation_id":"9b58cfe1-7e13-4a3b-a5d3-7ac32affe3c3","resolution":{"observed_at":"2026-07-02T12:46:57.283555Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2602.20021","last_updated":"2026-02-23T16:28:48Z","snapshot_observed_at":"2026-07-06T22:46:45.213871Z","submitted_at":"2026-02-23T16:28:48Z","title":"Agents of Chaos","version":1},"cited_work":{"arxiv_id":"2602.20021","doi":"10.48550/arxiv.2602.20021","metadata_source":"pith","pith_arxiv_id":"2602.20021","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Agents of Chaos","venue":"cs.AI","work_id":"02176c31-e67c-46e2-835a-ac21f005e1be","year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2602.20021","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:705a7b65a3a544668b4ec231fe2a16cebbf0922c4ad42bada1bd529ac71a6956","observation_id":"3849e3d0-15e6-4617-8d9b-78cc4f217974","resolution":{"observed_at":"2026-07-02T12:46:57.288595Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-07-15T18:20:21.538265+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-15T18:20:21.538265+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"ICLR 2025 Workshop on Building Trust in Language Models and Applications , year=","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:99e4c137a653e0c6266493f82a54e0179aee21ff237be330419840a447a75743","observation_id":"04ee2710-624d-4d15-a8ed-22a4121c84ff","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems , pages=","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:4acc8aa74c7c368a2609cefdbdeddc752fc6f303858aa56f67418454db8aea2c","observation_id":"724378d9-82a6-4909-9019-3b82872b7f95","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2601.21112","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-02T12:46:57.276653Z","title":"How Does Information Access Affect LLM Monitors’ Ability to Detect Sabotage?","venue":null,"work_id":"badb5d9d-46fe-4599-96aa-79a8b2c3b4c4","year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:d97f4aff7a8266f07ead745bcd65923b2ceebaa91bb2691aca67f443b5018148","observation_id":"f27775fe-bfa1-46bc-b86e-09ebca92eac6","resolution":{"observed_at":"2026-07-02T12:46:57.278635Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2401.05566","last_updated":"2024-01-17T20:26:01Z","snapshot_observed_at":"2026-07-06T17:14:03.152949Z","submitted_at":"2024-01-10T22:14:35Z","title":"Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training","version":3},"cited_work":{"arxiv_id":"2401.05566","doi":"10.48550/arxiv.2401.05566","metadata_source":"pith","pith_arxiv_id":"2401.05566","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training","venue":"cs.CR","work_id":"b95e7447-320c-4c85-b5d0-3708cc2cc72e","year":2024},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2401.05566","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:0081bbcd749667d29c1c18a7db9c22a5a6a6bc5a956ccd19216667f0848f1427","observation_id":"548eac33-5415-4f56-bcf3-012132e23035","resolution":{"observed_at":"2026-07-02T12:46:57.277907Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-01T07:38:13.966538+00:00","source":"crossref_status_cache"},{"observed_at":"2026-08-01T07:38:13.966538+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:76c110355d946ccb4166c742f50f36e1afbf77ae6d0b000d1a3cd5e3ea148fd4","observation_id":"2e3f7154-9cf2-4915-a185-ecdd0f8137fc","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"A Survey on Trustworthy","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:9bca934a43850c88443915a7def581bd3e6ef01c8f5e8470cd85ccea7f032265","observation_id":"e2308a4c-3a33-43ff-b475-5079dedeaeb2","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:9b8d872100ac8d9704e1397acd5bd5cc9e31312ed1bcd2286fd0f2c969bbfb25","observation_id":"a89f16ad-0224-4d49-8727-66d5f4a3f87e","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:5a4fcf436aeede8e4e387f65e43698236505d71ea154fc51bfc973fc51c019ed","observation_id":"5212ba0f-3037-41c2-86cc-a141e13a1a44","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:04b22c36a3172187e6f572d0567d6cce69d9fd42db398006f93ff3b5fad3366a","observation_id":"88c25f8e-ebc2-4222-bd8b-e495a781d2ea","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2511.09904","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-02T17:07:12.977337Z","title":"2025 , journal =","venue":null,"work_id":"b7f90ddb-4e1e-4425-a4a1-972f908a2c41","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:97f44e1e334b4284f84a4325a5124759f790e43c9febdc98f28c3cc2900819fb","observation_id":"158293f9-b2be-4a73-a15b-3f0466147936","resolution":{"observed_at":"2026-07-02T12:46:57.286024Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Async Control: Stress-Testing Asynchronous Control Measures for","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:68c6a9064d1911dfd0f4a813fa37aa35990317c237f70edd0e158647e3eebe24","observation_id":"c5a38714-17ac-4b11-80ed-2ca7e4a0e4cb","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19461","last_updated":"2025-08-26T22:29:31Z","snapshot_observed_at":"2026-08-07T19:17:20.900756Z","submitted_at":"2025-08-26T22:29:31Z","title":"Reliable Weak-to-Strong Monitoring of LLM Agents","version":1},"cited_work":{"arxiv_id":"2508.19461","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2508.19461","snapshot_observed_at":"2026-07-02T12:46:57.289769Z","title":"Reliable Weak-to-Strong Monitoring of","venue":null,"work_id":"5126ebcc-f0ae-48e2-8202-252206486c54","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2508.19461","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:bf0c0e119331f07bed1134ddd9cd204ddf27fa0135c09a5846bb6486432e069a","observation_id":"7da168a9-a7f1-4ba9-97e3-4cedd4cbf827","resolution":{"observed_at":"2026-07-02T12:46:57.291196Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2511.02823","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-02T12:46:57.263354Z","title":"2025 , journal =","venue":null,"work_id":"974145e1-204c-4505-ac87-209cfe052396","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:6ff6a9da4dff676150b18d145939adaab901a06d2c8378d620e99e6f99e1c338","observation_id":"aec8c6f0-c3d2-4711-994f-4b029290bde2","resolution":{"observed_at":"2026-07-02T12:46:57.264776Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2412.14093","last_updated":"2024-12-20T02:22:19Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-12-18T17:41:24Z","title":"Alignment faking in large language models","version":2},"cited_work":{"arxiv_id":"2412.14093","doi":"10.48550/arxiv.2412.14093","metadata_source":"pith","pith_arxiv_id":"2412.14093","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Alignment faking in large language models","venue":"cs.AI","work_id":"cc253a89-cda1-4889-9631-bf3ce8147650","year":2024},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2412.14093","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:9a257971e91a4d9893fe8f2702ac4531f5df8058c39a34c1b3e61eee7728e2e5","observation_id":"b61a4d78-074f-48f1-8ec3-94d54cc51402","resolution":{"observed_at":"2026-07-02T12:46:57.262186Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-01T07:38:13.802289+00:00","source":"crossref_status_cache"},{"observed_at":"2026-08-01T07:38:13.802289+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:e606c8d47ef2a63fac4cd4bb8190ed22a5cbd697e4578e8313b4ebd17df40056","observation_id":"6e3e06bf-5acd-4d39-b3e0-d68820c2c69b","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2602.13379","last_updated":"2026-06-10T21:28:47Z","snapshot_observed_at":"2026-08-08T14:12:41.932086Z","submitted_at":"2026-02-13T18:38:18Z","title":"Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents","version":2},"cited_work":{"arxiv_id":"2602.13379","doi":"10.48550/arxiv.2602.13379","metadata_source":"pith","pith_arxiv_id":"2602.13379","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Unsafer in many turns: Benchmarking and defending multi-turn safety risks in tool-using agents","venue":"cs.CR","work_id":"23cffd44-840b-4aca-827b-efc3ed0ec254","year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2602.13379","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:7d05fad8687b667b51fa47c2eed67f6a88b477a65d533ef7dd72804f1de5e68d","observation_id":"0715d350-1acc-43b0-a23c-ce60b0cebc9a","resolution":{"observed_at":"2026-07-02T12:46:57.267236Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Assessing","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:22b506c1a5255d644b337b3db92a933721e027ffba2c2d8e90362636c7e4e8b8","observation_id":"0fcbb8f8-b50b-4ca2-8579-d9b1692feb33","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2510.05179","doi":"10.48550/arxiv.2510.05179","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Ritchie, Soren Mindermann, Evan Hubinger, Ethan Perez, and Kevin Troy","venue":"ArXiv.org","work_id":"8fade7d3-e982-4854-a5f6-e1f8215b6722","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:a0558b4d86b68af7af33e98913222c5ec58bed098d9cc6c923b90506e9262353","observation_id":"f9e179ad-05c4-47b9-b70a-252c895875c4","resolution":{"observed_at":"2026-07-02T12:46:57.269807Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Natural Emergent Misalignment from Reward Hacking in Production","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:547adbaa3dc13c06b32387e67fff65ffb6e1f196cf04c0b5855465ecb617c560","observation_id":"4371d167-04c8-4802-bea1-f30afd6ec100","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2509.15541","doi":"10.48550/arxiv.2509.15541","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Stress Testing Deliberative Alignment for Anti-Scheming Training , url =","venue":"ArXiv.org","work_id":"c7e5c203-4279-4bcc-9d8f-65ca27bb5dd4","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:ebf5ccc480b587de0e26b23ad76e8a07deda7579f471b21ee67d5b5576ec7389","observation_id":"06f27233-de57-462a-bd99-04576628c8c3","resolution":{"observed_at":"2026-07-02T12:46:57.259807Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-07-15T18:20:38.241743+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-15T18:20:38.241743+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":", journal =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:1b55ad4c4e916c3d16308b4e8a372d478d621fef2d48abde10038fb24c2c23e2","observation_id":"73d7b82c-4499-4dc7-8ff0-f4c9ac8f398a","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2505.23575","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-10T00:56:41.071439Z","title":"Cot red-handed: Stress testing chain- of-thought monitoring.ArXiv, abs/2505.23575","venue":null,"work_id":"7b41fc27-304c-4254-9d77-c33e2a98d2d6","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:7200b9afd779790f738a58a7b0a5c3a7bfa82f53062a76c282ae9b9d59ef5d09","observation_id":"ce3459a6-c672-4bbe-a28f-7c0f62a61807","resolution":{"observed_at":"2026-07-02T12:46:57.253840Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"When Developer Aid Becomes Security Debt: A Systematic Analysis of Insecure Behaviors in","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:0e9f3108f291b63fd6fda700479cb2000b1b79ed9785c7b8b7ad45e66ebbfce7","observation_id":"5a8ca333-d64e-4da0-8a1b-3c54ac3a2859","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2510.05159","last_updated":"2026-08-03T18:11:09Z","snapshot_observed_at":"2026-08-07T23:09:07.293633Z","submitted_at":"2025-10-03T12:47:21Z","title":"Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chain","version":5},"cited_work":{"arxiv_id":"2510.05159","doi":null,"metadata_source":"pith","pith_arxiv_id":"2510.05159","snapshot_observed_at":"2026-07-02T12:46:57.246863Z","title":"Malice in Agentland: Down the Rabbit Hole of Backdoors in the AI Supply Chain","venue":"cs.CR","work_id":"86c39fa3-5597-4c4c-98fd-e688044242df","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2510.05159","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:88b6e2d959f25139fd827c3640b74b3e9723561a5cac35848c46080895d64775","observation_id":"f9adf27e-512e-4567-acbb-1d32dd378503","resolution":{"observed_at":"2026-07-02T12:46:57.248117Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Is Vibe Coding Safe?","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:5631cc82a25cf20d264d41c95fefc28ab94caa3c0817298247acd567d7177934","observation_id":"562f5cc0-a16f-41c5-b069-98ae9fa55e2c","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2601.17548","doi":"10.48550/arxiv.2601.17548","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Maloyan and D","venue":"CyberLeninK (CyberLeninka)","work_id":"16c69ef7-00fc-4f32-bd01-e17005fa7480","year":2026},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:b13ff47a508cca19c998dc9acb0867ae5b6e99c3363fd60fb0ee07394751b7e2","observation_id":"d336efa8-bfa0-4201-95aa-e1fe95390763","resolution":{"observed_at":"2026-07-02T12:46:57.293745Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Evolution of Programmers' Trust in Generative","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:8221b05d982a261070457d495b871934669c39b9fd49397ed6e6f08b147c529b","observation_id":"e7d06ee2-51e1-4aed-b1ab-d7a163826097","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Privacy Leakage Overshadowed by Views of","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:8332a728a9882d3af534f905b2e390ce0409626da184f87c212d6664edbe2577","observation_id":"73db284a-fb6e-4af1-9535-ac8873cb59d4","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Not What You've Signed Up For: Compromising Real-World","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:d8378b5c1c9f99e064343804fc04d0cac4bbe4e4dce3e34b93c92a0f8fa4840d","observation_id":"49fa82c2-df53-4912-87c2-0ec106798590","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.04984","last_updated":"2025-01-14T20:16:01Z","snapshot_observed_at":"2026-08-09T13:05:34.738390Z","submitted_at":"2024-12-06T12:09:50Z","title":"Frontier Models are Capable of In-context Scheming","version":2},"cited_work":{"arxiv_id":"2412.04984","doi":"10.48550/arxiv.2412.04984","metadata_source":"pith","pith_arxiv_id":"2412.04984","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Frontier Models are Capable of In-context Scheming","venue":"cs.AI","work_id":"1372f9da-8fac-4446-ba43-4e9e053c8b28","year":2024},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2412.04984","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:43bf8e499489441995a8c744375a1edbf146db752e4f873d174c27a98e08d48a","observation_id":"7a6e55f6-a435-42f7-9429-14e3cbb5b190","resolution":{"observed_at":"2026-07-02T12:46:57.251299Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-01T07:38:14.128935+00:00","source":"crossref_status_cache"},{"observed_at":"2026-08-01T07:38:14.128935+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2406.10162","last_updated":"2024-06-29T00:28:47Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-06-14T16:26:20Z","title":"Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models","version":3},"cited_work":{"arxiv_id":"2406.10162","doi":"10.48550/arxiv.2406.10162","metadata_source":"pith","pith_arxiv_id":"2406.10162","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models","venue":"cs.AI","work_id":"014812eb-baf1-4420-a49e-8896a973e595","year":2024},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2406.10162","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:9637a3644c9c9f6fa1a86b29f7c06cf5c6669b47f211c2c22e3b33bc8f0178be","observation_id":"fb42a312-2899-4dcb-b526-a34d427c02b4","resolution":{"observed_at":"2026-07-02T12:46:57.256904Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Asleep at the Keyboard?","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:07092b6878230677526e7fde3ed54c9d3f3c56623bb864334a899228e7a3ec20","observation_id":"c970b384-e342-48b3-9543-cc33e94b987e","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Do Users Write More Insecure Code with","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:af208d7b19a5020f833eed4c7ec107c8cd8509b0f9e26c4c6a745a6ebc149ac4","observation_id":"3bf517bd-ef18-433c-bd76-7d65558826ff","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:a369ce1896ace20da072a53ed9767d387b4b443887a4c98ac35f02c447f6d150","observation_id":"718e5478-0239-4c92-a815-0fedbc601566","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:a96f0fa24974c5925479efe940473d379d5bbee9cbe15537f3fff9c37a9a8794","observation_id":"b61d1e8d-b559-4a7e-ad9c-761dd17e43f1","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Red-Teaming","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:867448763f8567b14cc0bd9f7f013afcd49f00ce5efe97a3ba36744b19da8936","observation_id":"0e5fd8cd-7c50-489c-8cc6-e171ff683a6e","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:bc98156b53fa786db1e718f26aa839bd909aaa8f5acdcd6f930cb7378612a1f0","observation_id":"87090d60-508a-47b2-9f36-0551a4e07d2d","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.04957","last_updated":"2025-03-06T20:43:14Z","snapshot_observed_at":"2026-08-10T13:31:38.811194Z","submitted_at":"2025-03-06T20:43:14Z","title":"SafeArena: Evaluating the Safety of Autonomous Web Agents","version":1},"cited_work":{"arxiv_id":"2503.04957","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2503.04957","snapshot_observed_at":"2026-07-04T10:59:46.546058Z","title":"Safearena: Evaluating the safety of autonomous web agents","venue":null,"work_id":"4112362f-3fda-4a14-a761-3d5567824aa9","year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2503.04957","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:84678d5fe8e6fe06b3af5aa9fce937a456fc8113ed99c3450fbdbb8101d1677e","observation_id":"e11dc3fa-c38d-40ae-a0d0-364508b3e5ec","resolution":{"observed_at":"2026-07-02T12:46:57.272709Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Usability evaluation in industry , volume=","venue":null,"work_id":null,"year":1996},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:6354e8e83a9a7b77effc6e939af4bbe00c1350802633e2e39837ea024c907994","observation_id":"23b73752-0f46-4c3a-93ef-24e996d857f1","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.21514","last_updated":"2024-10-28T20:34:51Z","snapshot_observed_at":"2026-07-06T19:41:10.332290Z","submitted_at":"2024-10-28T20:34:51Z","title":"Sabotage Evaluations for Frontier Models","version":1},"cited_work":{"arxiv_id":"2410.21514","doi":"10.48550/arxiv.2410.21514","metadata_source":"arxiv_reference","pith_arxiv_id":"2410.21514","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"and Duvenaud, David , urldate =","venue":"arXiv (Cornell University)","work_id":"695ec6fa-c23c-4b40-b4ef-1d6a6d5107bc","year":2024},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"cited_paper":"/paper/2410.21514","citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:9dc3d2d429a7cf8dda2ca009e56ae91d3f8873160107d49ee49ae9b40227276d","observation_id":"28eb45e3-cc5b-416f-afe9-da1d99c19e17","resolution":{"observed_at":"2026-07-02T12:46:57.283112Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-07-15T10:49:59.945692+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-15T10:49:59.945692+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"Authorea Preprints , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:038a6198d947984dad864773f8428b96828609811273bf1bfdcf6598b55a30ec","observation_id":"92df3bc5-d35e-492c-a519-d8c4b080cffe","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-28T01:48:56.367899Z","title":"IEEE Access , volume=","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?","version":1},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-06-28T01:48:56.367899Z"},"links":{"citing_paper":"/paper/2606.05647"},"observation_digest":"sha256:40c4658a91d712e90b96563cf44e2581eecdab8ea6122b93e687fefbb0b68a8c","observation_id":"bb7f1493-42fe-4fb4-8f06-7c2896a09bd2","resolution":{"observed_at":"2026-06-28T01:48:56.367899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2606.05647","last_updated":"2026-06-04T03:22:17Z","latest_version":1,"primary_category":"cs.AI","snapshot_observed_at":"2026-08-01T22:43:20.237459Z","submitted_at":"2026-06-04T03:22:17Z","title":"Coding with \"Enemy\": Can Human Developers Detect AI Agent Sabotage?"},"reference_resolution":{"displayed":50,"state_counts":{"malformed_identifier":0,"metadata_mismatch":16,"parse_uncertain":0,"unresolved":29,"verified_exact":5,"verified_fuzzy":0},"total_outbound_references":50},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"thesis":"As of 11 August 2026, this Paper Citation Record lists 50 of 50 outbound references and 1 inbound Pith citation observation for arXiv:2606.05647."}