{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SB3GVF4QCT2JLWDVU7JJJHLIQN","short_pith_number":"pith:SB3GVF4Q","schema_version":"1.0","canonical_sha256":"90766a979014f495d875a7d2949d688360425564fb98caee2d2955f7647dbfaa","source":{"kind":"arxiv","id":"2605.27117","version":1},"attestation_state":"computed","paper":{"title":"Position: AI Safety Requires Effective Controllability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jun Sun, Yige Li, Yunhao Feng","submitted_at":"2026-05-26T14:53:24Z","abstract_excerpt":"AI safety is still largely framed as alignment: training models to follow human preferences, safety policies, and normative constraints. That framing has improved the behavior of modern language models, but aligned behavior does not by itself guarantee that a deployed agent can be stopped, overridden, or constrained once it operates in open-ended, interactive, and tool-using environments. A system may be safe in expectation and still fail to yield to explicit runtime authority under conflicting instructions, long-horizon execution, adversarial inputs, or risky tool use. This position paper arg"},"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":"2605.27117","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-26T14:53:24Z","cross_cats_sorted":[],"title_canon_sha256":"ba99537106da7860289f3d2876a6469c4934e27fd2f3579728c923569de1697f","abstract_canon_sha256":"5ec4b0a1de4697674f91311f9598ff16e719f401ab73dbf41fb5c4d1c51d98bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T02:05:42.825643Z","signature_b64":"WA6F0OvOF84LZiluIAgYJzXREUeRPGyymuoTcL2kb/TCBYQwH5ahmZvnR5P8MQcLhRHwPJG4L+tWsL3CsraaAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90766a979014f495d875a7d2949d688360425564fb98caee2d2955f7647dbfaa","last_reissued_at":"2026-05-27T02:05:42.824864Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T02:05:42.824864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Position: AI Safety Requires Effective Controllability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jun Sun, Yige Li, Yunhao Feng","submitted_at":"2026-05-26T14:53:24Z","abstract_excerpt":"AI safety is still largely framed as alignment: training models to follow human preferences, safety policies, and normative constraints. That framing has improved the behavior of modern language models, but aligned behavior does not by itself guarantee that a deployed agent can be stopped, overridden, or constrained once it operates in open-ended, interactive, and tool-using environments. A system may be safe in expectation and still fail to yield to explicit runtime authority under conflicting instructions, long-horizon execution, adversarial inputs, or risky tool use. This position paper arg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27117","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/2605.27117/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":"2605.27117","created_at":"2026-05-27T02:05:42.825000+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27117v1","created_at":"2026-05-27T02:05:42.825000+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27117","created_at":"2026-05-27T02:05:42.825000+00:00"},{"alias_kind":"pith_short_12","alias_value":"SB3GVF4QCT2J","created_at":"2026-05-27T02:05:42.825000+00:00"},{"alias_kind":"pith_short_16","alias_value":"SB3GVF4QCT2JLWDV","created_at":"2026-05-27T02:05:42.825000+00:00"},{"alias_kind":"pith_short_8","alias_value":"SB3GVF4Q","created_at":"2026-05-27T02:05:42.825000+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN","json":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN.json","graph_json":"https://pith.science/api/pith-number/SB3GVF4QCT2JLWDVU7JJJHLIQN/graph.json","events_json":"https://pith.science/api/pith-number/SB3GVF4QCT2JLWDVU7JJJHLIQN/events.json","paper":"https://pith.science/paper/SB3GVF4Q"},"agent_actions":{"view_html":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN","download_json":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN.json","view_paper":"https://pith.science/paper/SB3GVF4Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27117&json=true","fetch_graph":"https://pith.science/api/pith-number/SB3GVF4QCT2JLWDVU7JJJHLIQN/graph.json","fetch_events":"https://pith.science/api/pith-number/SB3GVF4QCT2JLWDVU7JJJHLIQN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN/action/storage_attestation","attest_author":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN/action/author_attestation","sign_citation":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN/action/citation_signature","submit_replication":"https://pith.science/pith/SB3GVF4QCT2JLWDVU7JJJHLIQN/action/replication_record"}},"created_at":"2026-05-27T02:05:42.825000+00:00","updated_at":"2026-05-27T02:05:42.825000+00:00"}