{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:B2PLYHCB7Q7YCDMWJDV7YSJTLR","short_pith_number":"pith:B2PLYHCB","schema_version":"1.0","canonical_sha256":"0e9ebc1c41fc3f810d9648ebfc49335c5c0ce513c0c090a8d3a0120281be82d6","source":{"kind":"arxiv","id":"2506.23844","version":1},"attestation_state":"computed","paper":{"title":"A Survey on Autonomy-Induced Security Risks in Large Model-Based Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chang Liu, Hang Su, Jun Luo, Jun Zhu, Xiao Yang, Yichi Zhang, Yinpeng Dong","submitted_at":"2025-06-30T13:34:34Z","abstract_excerpt":"Recent advances in large language models (LLMs) have catalyzed the rise of autonomous AI agents capable of perceiving, reasoning, and acting in dynamic, open-ended environments. These large-model agents mark a paradigm shift from static inference systems to interactive, memory-augmented entities. While these capabilities significantly expand the functional scope of AI, they also introduce qualitatively novel security risks - such as memory poisoning, tool misuse, reward hacking, and emergent misalignment - that extend beyond the threat models of conventional systems or standalone LLMs. In this"},"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":"2506.23844","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-30T13:34:34Z","cross_cats_sorted":[],"title_canon_sha256":"62ea61c8edd15be021d4a7e8c75760f1809f27afc8f77448854b7f33d3686f19","abstract_canon_sha256":"1997f311ac9045dd4acc05773152ea6b0f8144d45c29553b285712cc641a2744"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:29:28.819006Z","signature_b64":"9fieegX77ZN1R9ifLf8dhfxcN5Ty+2AABhZMGExzWg3HB7ZGW3HtD3auaX6mRQ0BcjUVyFAApc1djSgvhgdJDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e9ebc1c41fc3f810d9648ebfc49335c5c0ce513c0c090a8d3a0120281be82d6","last_reissued_at":"2026-07-05T11:29:28.818386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:29:28.818386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Survey on Autonomy-Induced Security Risks in Large Model-Based Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chang Liu, Hang Su, Jun Luo, Jun Zhu, Xiao Yang, Yichi Zhang, Yinpeng Dong","submitted_at":"2025-06-30T13:34:34Z","abstract_excerpt":"Recent advances in large language models (LLMs) have catalyzed the rise of autonomous AI agents capable of perceiving, reasoning, and acting in dynamic, open-ended environments. These large-model agents mark a paradigm shift from static inference systems to interactive, memory-augmented entities. While these capabilities significantly expand the functional scope of AI, they also introduce qualitatively novel security risks - such as memory poisoning, tool misuse, reward hacking, and emergent misalignment - that extend beyond the threat models of conventional systems or standalone LLMs. In this"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23844","kind":"arxiv","version":1},"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/2506.23844/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":"2506.23844","created_at":"2026-07-05T11:29:28.818482+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.23844v1","created_at":"2026-07-05T11:29:28.818482+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23844","created_at":"2026-07-05T11:29:28.818482+00:00"},{"alias_kind":"pith_short_12","alias_value":"B2PLYHCB7Q7Y","created_at":"2026-07-05T11:29:28.818482+00:00"},{"alias_kind":"pith_short_16","alias_value":"B2PLYHCB7Q7YCDMW","created_at":"2026-07-05T11:29:28.818482+00:00"},{"alias_kind":"pith_short_8","alias_value":"B2PLYHCB","created_at":"2026-07-05T11:29:28.818482+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01236","citing_title":"Safeguarding LLM Agents from Misalignment through Provenance Analysis","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10749","citing_title":"Toward Secure LLM Agents: Threat Surfaces, Attacks, Defenses, and Evaluation","ref_index":163,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22643","citing_title":"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23330","citing_title":"Security, Privacy, and Ethical Risks in OpenClaw","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22643","citing_title":"Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17830","citing_title":"Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01970","citing_title":"Trojan Hippo: Weaponizing Agent Memory for Data Exfiltration","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01970","citing_title":"Trojan Hippo: Weaponizing Agent Memory for Data Exfiltration","ref_index":84,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR","json":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR.json","graph_json":"https://pith.science/api/pith-number/B2PLYHCB7Q7YCDMWJDV7YSJTLR/graph.json","events_json":"https://pith.science/api/pith-number/B2PLYHCB7Q7YCDMWJDV7YSJTLR/events.json","paper":"https://pith.science/paper/B2PLYHCB"},"agent_actions":{"view_html":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR","download_json":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR.json","view_paper":"https://pith.science/paper/B2PLYHCB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.23844&json=true","fetch_graph":"https://pith.science/api/pith-number/B2PLYHCB7Q7YCDMWJDV7YSJTLR/graph.json","fetch_events":"https://pith.science/api/pith-number/B2PLYHCB7Q7YCDMWJDV7YSJTLR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR/action/storage_attestation","attest_author":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR/action/author_attestation","sign_citation":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR/action/citation_signature","submit_replication":"https://pith.science/pith/B2PLYHCB7Q7YCDMWJDV7YSJTLR/action/replication_record"}},"created_at":"2026-07-05T11:29:28.818482+00:00","updated_at":"2026-07-05T11:29:28.818482+00:00"}