{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:7E7IYP4PMSNHCEODMSD7XBHDN4","short_pith_number":"pith:7E7IYP4P","schema_version":"1.0","canonical_sha256":"f93e8c3f8f649a7111c36487fb84e36f296152c8447283cff168bfa5d5598dbc","source":{"kind":"arxiv","id":"2311.10538","version":3},"attestation_state":"computed","paper":{"title":"Testing Language Model Agents Safely in the Wild","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Adam Tauman Kalai, Craig Swift, David Atkinson, David Bau, Douglas Schonholtz, Marc Green, Merwane Hamadi, Silen Naihin","submitted_at":"2023-11-17T14:06:05Z","abstract_excerpt":"A prerequisite for safe autonomy-in-the-wild is safe testing-in-the-wild. Yet real-world autonomous tests face several unique safety challenges, both due to the possibility of causing harm during a test, as well as the risk of encountering new unsafe agent behavior through interactions with real-world and potentially malicious actors. We propose a framework for conducting safe autonomous agent tests on the open internet: agent actions are audited by a context-sensitive monitor that enforces a stringent safety boundary to stop an unsafe test, with suspect behavior ranked and logged to be examin"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2311.10538","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-11-17T14:06:05Z","cross_cats_sorted":[],"title_canon_sha256":"613e8dd59f6aa6b9b9f94069eabb6f8d321d9469b86d55c6e8d62ca682e72382","abstract_canon_sha256":"c287c96b37cca252143b68ef14337071d73cc1b24aa29e98118efab34ba75b9e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:19:23.959613Z","signature_b64":"EM/tDxZgZ6+YH9Gd2b2iLSllQCk+l4S9rj9DldWHCs2mwy7boeKHyGaJymY2sGbXE3k7DaQt+1WHAjUOfbu4Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f93e8c3f8f649a7111c36487fb84e36f296152c8447283cff168bfa5d5598dbc","last_reissued_at":"2026-07-05T07:19:23.959007Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:19:23.959007Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Testing Language Model Agents Safely in the Wild","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Adam Tauman Kalai, Craig Swift, David Atkinson, David Bau, Douglas Schonholtz, Marc Green, Merwane Hamadi, Silen Naihin","submitted_at":"2023-11-17T14:06:05Z","abstract_excerpt":"A prerequisite for safe autonomy-in-the-wild is safe testing-in-the-wild. Yet real-world autonomous tests face several unique safety challenges, both due to the possibility of causing harm during a test, as well as the risk of encountering new unsafe agent behavior through interactions with real-world and potentially malicious actors. We propose a framework for conducting safe autonomous agent tests on the open internet: agent actions are audited by a context-sensitive monitor that enforces a stringent safety boundary to stop an unsafe test, with suspect behavior ranked and logged to be examin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.10538","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2311.10538/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2311.10538","created_at":"2026-07-05T07:19:23.959069+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.10538v3","created_at":"2026-07-05T07:19:23.959069+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.10538","created_at":"2026-07-05T07:19:23.959069+00:00"},{"alias_kind":"pith_short_12","alias_value":"7E7IYP4PMSNH","created_at":"2026-07-05T07:19:23.959069+00:00"},{"alias_kind":"pith_short_16","alias_value":"7E7IYP4PMSNHCEOD","created_at":"2026-07-05T07:19:23.959069+00:00"},{"alias_kind":"pith_short_8","alias_value":"7E7IYP4P","created_at":"2026-07-05T07:19:23.959069+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14133","citing_title":"ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2512.08104","citing_title":"AgentCrypt: Advancing Privacy and (Secure) Computation in AI Agent Collaboration","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14133","citing_title":"ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09163","citing_title":"FORTIS: Benchmarking Over-Privilege in Agent Skills","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09163","citing_title":"FORTIS: Benchmarking Over-Privilege in Agent Skills","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4","json":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4.json","graph_json":"https://pith.science/api/pith-number/7E7IYP4PMSNHCEODMSD7XBHDN4/graph.json","events_json":"https://pith.science/api/pith-number/7E7IYP4PMSNHCEODMSD7XBHDN4/events.json","paper":"https://pith.science/paper/7E7IYP4P"},"agent_actions":{"view_html":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4","download_json":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4.json","view_paper":"https://pith.science/paper/7E7IYP4P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.10538&json=true","fetch_graph":"https://pith.science/api/pith-number/7E7IYP4PMSNHCEODMSD7XBHDN4/graph.json","fetch_events":"https://pith.science/api/pith-number/7E7IYP4PMSNHCEODMSD7XBHDN4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4/action/storage_attestation","attest_author":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4/action/author_attestation","sign_citation":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4/action/citation_signature","submit_replication":"https://pith.science/pith/7E7IYP4PMSNHCEODMSD7XBHDN4/action/replication_record"}},"created_at":"2026-07-05T07:19:23.959069+00:00","updated_at":"2026-07-05T07:19:23.959069+00:00"}