{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:WQYYZWOSJCIA2YBDSNWAE625C4","short_pith_number":"pith:WQYYZWOS","schema_version":"1.0","canonical_sha256":"b4318cd9d248900d6023936c027b5d17191b0511dedca52c0b2ab44407411e96","source":{"kind":"arxiv","id":"1802.07228","version":2},"attestation_state":"computed","paper":{"title":"The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CY"],"primary_cat":"cs.AI","authors_text":"Allan Dafoe, Ben Garfinkel, Bobby Filar, Carrick Flynn, Clare Lyle, Dario Amodei, Gregory C. Allen, Haydn Belfield, Heather Roff, Helen Toner, Hyrum Anderson, Jack Clark, Jacob Steinhardt, Joanna Bryson, Michael Page, Miles Brundage, Owain Evans, Paul Scharre, Peter Eckersley, Rebecca Crootof, Roman Yampolskiy, Se\\'an \\'O h\\'Eigeartaigh, Sebastian Farquhar, Shahar Avin, SJ Beard, Thomas Zeitzoff","submitted_at":"2018-02-20T18:07:50Z","abstract_excerpt":"This report surveys the landscape of potential security threats from malicious uses of AI, and proposes ways to better forecast, prevent, and mitigate these threats. After analyzing the ways in which AI may influence the threat landscape in the digital, physical, and political domains, we make four high-level recommendations for AI researchers and other stakeholders. We also suggest several promising areas for further research that could expand the portfolio of defenses, or make attacks less effective or harder to execute. Finally, we discuss, but do not conclusively resolve, the long-term equ"},"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":"1802.07228","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2018-02-20T18:07:50Z","cross_cats_sorted":["cs.CR","cs.CY"],"title_canon_sha256":"9db3a540be8f7893d94761a5e611855eff1c9eb49e4918b74eb75347c3648ba3","abstract_canon_sha256":"39f9050e1017930f3898f1fc70a2aa322eff144ac6142e7709dda18f13fe6d9d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:42:16.078304Z","signature_b64":"1k7TXw9fBtcvfTQ4ytEQ8+HgzlWNH+Ng2/JE2s+3URKj+vxTM8ZMSXiN2XOpRiTnETjsx+g8De/GYfuolDF0Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4318cd9d248900d6023936c027b5d17191b0511dedca52c0b2ab44407411e96","last_reissued_at":"2026-07-05T09:42:16.077792Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:42:16.077792Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CY"],"primary_cat":"cs.AI","authors_text":"Allan Dafoe, Ben Garfinkel, Bobby Filar, Carrick Flynn, Clare Lyle, Dario Amodei, Gregory C. Allen, Haydn Belfield, Heather Roff, Helen Toner, Hyrum Anderson, Jack Clark, Jacob Steinhardt, Joanna Bryson, Michael Page, Miles Brundage, Owain Evans, Paul Scharre, Peter Eckersley, Rebecca Crootof, Roman Yampolskiy, Se\\'an \\'O h\\'Eigeartaigh, Sebastian Farquhar, Shahar Avin, SJ Beard, Thomas Zeitzoff","submitted_at":"2018-02-20T18:07:50Z","abstract_excerpt":"This report surveys the landscape of potential security threats from malicious uses of AI, and proposes ways to better forecast, prevent, and mitigate these threats. After analyzing the ways in which AI may influence the threat landscape in the digital, physical, and political domains, we make four high-level recommendations for AI researchers and other stakeholders. We also suggest several promising areas for further research that could expand the portfolio of defenses, or make attacks less effective or harder to execute. Finally, we discuss, but do not conclusively resolve, the long-term equ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1802.07228","kind":"arxiv","version":2},"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/1802.07228/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":"1802.07228","created_at":"2026-07-05T09:42:16.077853+00:00"},{"alias_kind":"arxiv_version","alias_value":"1802.07228v2","created_at":"2026-07-05T09:42:16.077853+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1802.07228","created_at":"2026-07-05T09:42:16.077853+00:00"},{"alias_kind":"pith_short_12","alias_value":"WQYYZWOSJCIA","created_at":"2026-07-05T09:42:16.077853+00:00"},{"alias_kind":"pith_short_16","alias_value":"WQYYZWOSJCIA2YBD","created_at":"2026-07-05T09:42:16.077853+00:00"},{"alias_kind":"pith_short_8","alias_value":"WQYYZWOS","created_at":"2026-07-05T09:42:16.077853+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":24,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22875","citing_title":"FedOT: Ownership Verification and Leakage Tracing via Watermarks for Federated LDMs","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22237","citing_title":"Investigating The Security of Modern AI and Cloud Infrastructure","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02144","citing_title":"Taxing Artificial Intelligence","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30650","citing_title":"When AI Meets Wall Street: A Survey on Trustworthy AI in Fintech","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"1906.08743","citing_title":"We Need No Pixels: Video Manipulation Detection Using Stream Descriptors","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"1907.04534","citing_title":"The Role of Cooperation in Responsible AI Development","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"1907.11274","citing_title":"Reducing malicious use of synthetic media research: Considerations and potential release practices for machine learning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2601.06163","citing_title":"Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2603.10015","citing_title":"The coordination gap in frontier AI safety policies","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16589","citing_title":"STRIKE: A Structured Taxonomy of Cybercrime for Risk, Impact, Knowledge, and Evolution","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2506.07586","citing_title":"MalGEN: A Testbed for Modeling and Evaluating Malware Behaviors","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2404.16014","citing_title":"Improving Dictionary Learning with Gated Sparse Autoencoders","ref_index":204,"is_internal_anchor":false},{"citing_arxiv_id":"1909.05858","citing_title":"CTRL: A Conditional Transformer Language Model for Controllable Generation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2109.13916","citing_title":"Unsolved Problems in ML Safety","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20867","citing_title":"Preserving Decision Sovereignty in Military AI: A Trade-Secret-Safe Architectural Framework for Model Replaceability, Human Authority, and State Control","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26167","citing_title":"Gaussian Shannon: High-Precision Diffusion Model Watermarking Based on Communication","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24794","citing_title":"V.O.I.C.E (Voice, Ownership, Identity, Control, Expression): Risk Taxonomy of Synthetic Voice Generation From Empirical Data","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01415","citing_title":"AI Safety as Control of Irreversibility: A Systems Framework for Decision-Energy and Sovereignty Boundaries","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20300","citing_title":"FSFM: A Biologically-Inspired Framework for Selective Forgetting of Agent Memory","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10460","citing_title":"Toward Accountable AI-Generated Content on Social Platforms: Steganographic Attribution and Multimodal Harm Detection","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09544","citing_title":"Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07838","citing_title":"We Need Strong Preconditions For Using Simulations In Policy","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2101.00027","citing_title":"The Pile: An 800GB Dataset of Diverse Text for Language Modeling","ref_index":186,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19354","citing_title":"Do Agents Dream of Root Shells? Partial-Credit Evaluation of LLM Agents in Capture the Flag Challenges","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4","json":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4.json","graph_json":"https://pith.science/api/pith-number/WQYYZWOSJCIA2YBDSNWAE625C4/graph.json","events_json":"https://pith.science/api/pith-number/WQYYZWOSJCIA2YBDSNWAE625C4/events.json","paper":"https://pith.science/paper/WQYYZWOS"},"agent_actions":{"view_html":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4","download_json":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4.json","view_paper":"https://pith.science/paper/WQYYZWOS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1802.07228&json=true","fetch_graph":"https://pith.science/api/pith-number/WQYYZWOSJCIA2YBDSNWAE625C4/graph.json","fetch_events":"https://pith.science/api/pith-number/WQYYZWOSJCIA2YBDSNWAE625C4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4/action/storage_attestation","attest_author":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4/action/author_attestation","sign_citation":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4/action/citation_signature","submit_replication":"https://pith.science/pith/WQYYZWOSJCIA2YBDSNWAE625C4/action/replication_record"}},"created_at":"2026-07-05T09:42:16.077853+00:00","updated_at":"2026-07-05T09:42:16.077853+00:00"}