{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:B2AMVS6C3DNOQQUHG6CFMKXX4R","short_pith_number":"pith:B2AMVS6C","schema_version":"1.0","canonical_sha256":"0e80cacbc2d8dae842873784562af7e454a9265b5860823ff7c719f383604669","source":{"kind":"arxiv","id":"2601.06366","version":2},"attestation_state":"computed","paper":{"title":"SafeGPT: Preventing Data Leakage and Unethical Outputs in Enterprise LLM Use","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Luoxi Tang, Pratyush Desai, Yuqiao Meng, Zhaohan Xi","submitted_at":"2026-01-10T00:33:38Z","abstract_excerpt":"Large Language Models (LLMs) are transforming enterprise workflows but introduce security and ethics challenges when employees inadvertently share confidential data or generate policy-violating content. This paper proposes SafeGPT, a two-sided guardrail system preventing sensitive data leakage and unethical outputs. SafeGPT integrates input-side detection/redaction, output-side moderation/reframing, and human-in-the-loop feedback. Experiments demonstrate SafeGPT effectively reduces data leakage risk and biased outputs while maintaining satisfaction."},"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":"2601.06366","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-01-10T00:33:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d8a6ca82f8309b8746e4fb74895310d0bf412ce18adb132a6ab32cd73837ac74","abstract_canon_sha256":"175b43e7538d9b810122735f20486a9c3ce25204040f234b4c3092956f6166da"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:31.972106Z","signature_b64":"lHcPugjpK2NKfD7ozVwoZdDB0VHxB0jsBlw4G2JV+zdrWAv7SHaAZ/480IDL4NgpDibDCU0Akpc6qhK1yI8+Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e80cacbc2d8dae842873784562af7e454a9265b5860823ff7c719f383604669","last_reissued_at":"2026-05-20T00:00:31.971284Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:31.971284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SafeGPT: Preventing Data Leakage and Unethical Outputs in Enterprise LLM Use","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Luoxi Tang, Pratyush Desai, Yuqiao Meng, Zhaohan Xi","submitted_at":"2026-01-10T00:33:38Z","abstract_excerpt":"Large Language Models (LLMs) are transforming enterprise workflows but introduce security and ethics challenges when employees inadvertently share confidential data or generate policy-violating content. This paper proposes SafeGPT, a two-sided guardrail system preventing sensitive data leakage and unethical outputs. SafeGPT integrates input-side detection/redaction, output-side moderation/reframing, and human-in-the-loop feedback. Experiments demonstrate SafeGPT effectively reduces data leakage risk and biased outputs while maintaining satisfaction."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.06366","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/2601.06366/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":"2601.06366","created_at":"2026-05-20T00:00:31.971403+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.06366v2","created_at":"2026-05-20T00:00:31.971403+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.06366","created_at":"2026-05-20T00:00:31.971403+00:00"},{"alias_kind":"pith_short_12","alias_value":"B2AMVS6C3DNO","created_at":"2026-05-20T00:00:31.971403+00:00"},{"alias_kind":"pith_short_16","alias_value":"B2AMVS6C3DNOQQUH","created_at":"2026-05-20T00:00:31.971403+00:00"},{"alias_kind":"pith_short_8","alias_value":"B2AMVS6C","created_at":"2026-05-20T00:00:31.971403+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.08838","citing_title":"Generating Leakage-Free Benchmarks for Robust RAG Evaluation","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R","json":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R.json","graph_json":"https://pith.science/api/pith-number/B2AMVS6C3DNOQQUHG6CFMKXX4R/graph.json","events_json":"https://pith.science/api/pith-number/B2AMVS6C3DNOQQUHG6CFMKXX4R/events.json","paper":"https://pith.science/paper/B2AMVS6C"},"agent_actions":{"view_html":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R","download_json":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R.json","view_paper":"https://pith.science/paper/B2AMVS6C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.06366&json=true","fetch_graph":"https://pith.science/api/pith-number/B2AMVS6C3DNOQQUHG6CFMKXX4R/graph.json","fetch_events":"https://pith.science/api/pith-number/B2AMVS6C3DNOQQUHG6CFMKXX4R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R/action/storage_attestation","attest_author":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R/action/author_attestation","sign_citation":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R/action/citation_signature","submit_replication":"https://pith.science/pith/B2AMVS6C3DNOQQUHG6CFMKXX4R/action/replication_record"}},"created_at":"2026-05-20T00:00:31.971403+00:00","updated_at":"2026-05-20T00:00:31.971403+00:00"}