{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZBTRX2S2XORGMAW7Y3ZBFQTA5Q","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d88e00bb965c4981e7747a2135393161f81fcaf3f183f6941ec8c2a80052abd1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T17:56:18Z","title_canon_sha256":"095b150f2974f585d4ad341936cb4d8184351a5d45d7ea59aba7be3060369f6c"},"schema_version":"1.0","source":{"id":"2605.30322","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.30322","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.30322v1","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30322","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"pith_short_12","alias_value":"ZBTRX2S2XORG","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"pith_short_16","alias_value":"ZBTRX2S2XORGMAW7","created_at":"2026-05-29T02:06:16Z"},{"alias_kind":"pith_short_8","alias_value":"ZBTRX2S2","created_at":"2026-05-29T02:06:16Z"}],"graph_snapshots":[{"event_id":"sha256:e127616d88713a42060c12897bc206aca86c67f06b6a654002e134c4bccc4df7","target":"graph","created_at":"2026-05-29T02:06:16Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.30322/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce Gram, an automated alignment auditing framework to assess the propensity of AI agents to engage in sabotage. We evaluate Gemini models across 17 simulated agentic deployment scenarios that incentivize sabotage. We find Gemini models misbehave in about 2-3% of our simulated trajectories. Many of these cases are explained by \"overeagerness\" in Gemini models resulting in both excessive role-playing and goal-seeking behavior. In contrast to other alignment auditing approaches, Gram is designed to specifically evaluate misalignment and intentional sabotage in agentic coding and researc","authors_text":"David Lindner, Sebastian Farquhar, Victoria Krakovna","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T17:56:18Z","title":"Gram: Assessing sabotage propensities via automated alignment auditing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30322","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:61d6cc6ad6dc9e714c090b845e043d457e414a8102112e2de8ca5ef2d2545bac","target":"record","created_at":"2026-05-29T02:06:16Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d88e00bb965c4981e7747a2135393161f81fcaf3f183f6941ec8c2a80052abd1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T17:56:18Z","title_canon_sha256":"095b150f2974f585d4ad341936cb4d8184351a5d45d7ea59aba7be3060369f6c"},"schema_version":"1.0","source":{"id":"2605.30322","kind":"arxiv","version":1}},"canonical_sha256":"c8671bea5abba26602dfc6f212c260ec035dfaf84eb950d0b274660c296db85a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c8671bea5abba26602dfc6f212c260ec035dfaf84eb950d0b274660c296db85a","first_computed_at":"2026-05-29T02:06:16.480368Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T02:06:16.480368Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NULenM4+43WPAac7OlZTUrktmHZvsZ2WSc9HGMqFnZBk+pj3POvPv9yzN0IuLZcIYb7i9yH6Wzah6OvbamUNAw==","signature_status":"signed_v1","signed_at":"2026-05-29T02:06:16.480775Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.30322","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:61d6cc6ad6dc9e714c090b845e043d457e414a8102112e2de8ca5ef2d2545bac","sha256:e127616d88713a42060c12897bc206aca86c67f06b6a654002e134c4bccc4df7"],"state_sha256":"88f57d51b06c72a064dcf15d9b579bc0d996e1374a8f42fe44bc43da97d6d3e9"}