{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YCCDF6FS6BC5Q6ZUDLB34NKKPR","short_pith_number":"pith:YCCDF6FS","schema_version":"1.0","canonical_sha256":"c08432f8b2f045d87b341ac3be354a7c574e7caa3b0e19312652636d26c4239d","source":{"kind":"arxiv","id":"2505.08204","version":1},"attestation_state":"computed","paper":{"title":"LM-Scout: Analyzing the Security of Language Model Integration in Android Apps","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"(2) Google, (3) Purdue University), Antonio Bianchi (3) ((1) Georgia Institute of Technology, Aravind Machiry (3), G\\H{u}liz Seray Tuncay (2), Muhammad Ibrahim (1), Z. Berkay Celik (3)","submitted_at":"2025-05-13T03:37:23Z","abstract_excerpt":"Developers are increasingly integrating Language Models (LMs) into their mobile apps to provide features such as chat-based assistants. To prevent LM misuse, they impose various restrictions, including limits on the number of queries, input length, and allowed topics. However, if the LM integration is insecure, attackers can bypass these restrictions and gain unrestricted access to the LM, potentially harming developers' reputations and leading to significant financial losses.\n  This paper presents the first systematic study of insecure usage of LMs by Android apps. We first manually analyze a"},"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":"2505.08204","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-05-13T03:37:23Z","cross_cats_sorted":[],"title_canon_sha256":"d29ff012caa3feff117858b4d1b6c750acb109d955616b9e79c215ddf932ee06","abstract_canon_sha256":"62ef684d42e4de53b88fbc51b4500a151e28d3937fe342ca010aedf02fdfa453"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:26.465223Z","signature_b64":"xGYesbPvYLRhAfCB5rH1SOcS+vBNuQwagh48QuFkeiEVaR6lViWoRBXlDCYsRGmTJdEfONbr3MI+CdexKQbpDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c08432f8b2f045d87b341ac3be354a7c574e7caa3b0e19312652636d26c4239d","last_reissued_at":"2026-07-05T11:02:26.464748Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:26.464748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LM-Scout: Analyzing the Security of Language Model Integration in Android Apps","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"(2) Google, (3) Purdue University), Antonio Bianchi (3) ((1) Georgia Institute of Technology, Aravind Machiry (3), G\\H{u}liz Seray Tuncay (2), Muhammad Ibrahim (1), Z. Berkay Celik (3)","submitted_at":"2025-05-13T03:37:23Z","abstract_excerpt":"Developers are increasingly integrating Language Models (LMs) into their mobile apps to provide features such as chat-based assistants. To prevent LM misuse, they impose various restrictions, including limits on the number of queries, input length, and allowed topics. However, if the LM integration is insecure, attackers can bypass these restrictions and gain unrestricted access to the LM, potentially harming developers' reputations and leading to significant financial losses.\n  This paper presents the first systematic study of insecure usage of LMs by Android apps. We first manually analyze a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08204","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/2505.08204/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":"2505.08204","created_at":"2026-07-05T11:02:26.464813+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08204v1","created_at":"2026-07-05T11:02:26.464813+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08204","created_at":"2026-07-05T11:02:26.464813+00:00"},{"alias_kind":"pith_short_12","alias_value":"YCCDF6FS6BC5","created_at":"2026-07-05T11:02:26.464813+00:00"},{"alias_kind":"pith_short_16","alias_value":"YCCDF6FS6BC5Q6ZU","created_at":"2026-07-05T11:02:26.464813+00:00"},{"alias_kind":"pith_short_8","alias_value":"YCCDF6FS","created_at":"2026-07-05T11:02:26.464813+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12212","citing_title":"Mind your key: An Empirical Study of LLM API Credential Leakage in iOS Apps","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06095","citing_title":"LLM4CodeRE: Generative AI for Code Decompilation Analysis and Reverse Engineering","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR","json":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR.json","graph_json":"https://pith.science/api/pith-number/YCCDF6FS6BC5Q6ZUDLB34NKKPR/graph.json","events_json":"https://pith.science/api/pith-number/YCCDF6FS6BC5Q6ZUDLB34NKKPR/events.json","paper":"https://pith.science/paper/YCCDF6FS"},"agent_actions":{"view_html":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR","download_json":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR.json","view_paper":"https://pith.science/paper/YCCDF6FS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08204&json=true","fetch_graph":"https://pith.science/api/pith-number/YCCDF6FS6BC5Q6ZUDLB34NKKPR/graph.json","fetch_events":"https://pith.science/api/pith-number/YCCDF6FS6BC5Q6ZUDLB34NKKPR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR/action/storage_attestation","attest_author":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR/action/author_attestation","sign_citation":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR/action/citation_signature","submit_replication":"https://pith.science/pith/YCCDF6FS6BC5Q6ZUDLB34NKKPR/action/replication_record"}},"created_at":"2026-07-05T11:02:26.464813+00:00","updated_at":"2026-07-05T11:02:26.464813+00:00"}