{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:G4RM7W4NBCB3PUJAZN5TES6OP4","short_pith_number":"pith:G4RM7W4N","schema_version":"1.0","canonical_sha256":"3722cfdb8d0883b7d120cb7b324bce7f080c91dc6c0d3e3f4b60d2e45b3b4a22","source":{"kind":"arxiv","id":"2505.18953","version":2},"attestation_state":"computed","paper":{"title":"Evaluating AI for Finance: Is AI Credible at Assessing Investment Risk?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhinav Raghunathan, Ashita Bhutada, Cathy Guo, Dar Win Liew, Divij Chawla, Do Duc Anh, Prannaya Gupta, Rajat Bhardwaj, Rishabh Bhardwaj, Soujanya Poria, Vinod SP","submitted_at":"2025-05-25T02:56:19Z","abstract_excerpt":"We assess whether AI systems can credibly evaluate investment risk appetite-a task that must be thoroughly validated before automation. Our analysis was conducted on proprietary systems (GPT, Claude, Gemini) and open-weight models (LLaMA, DeepSeek, Mistral), using carefully curated user profiles that reflect real users with varying attributes such as country and gender. As a result, the models exhibit significant variance in score distributions when user attributes-such as country or gender-that should not influence risk computation are changed. For example, GPT-4o assigns higher risk scores t"},"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":"2505.18953","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-25T02:56:19Z","cross_cats_sorted":[],"title_canon_sha256":"2d5a8b01befcc4d480ed67c1db7e312727e4419089caaa8124bf7db4150b1c73","abstract_canon_sha256":"cccb69823f2759032afb0cb8b7183b6895da7cf10b3dc19e61f699fb3b38573b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:29.719045Z","signature_b64":"I37NbMthI6vX6VxrjO14/CflB7PcMdM48oryeqNPwv4wSdn5CdIWlTDaFSb2hzrGG6iLkjWA4hPOJUKeEk2lBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3722cfdb8d0883b7d120cb7b324bce7f080c91dc6c0d3e3f4b60d2e45b3b4a22","last_reissued_at":"2026-07-05T11:32:29.718558Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:29.718558Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating AI for Finance: Is AI Credible at Assessing Investment Risk?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhinav Raghunathan, Ashita Bhutada, Cathy Guo, Dar Win Liew, Divij Chawla, Do Duc Anh, Prannaya Gupta, Rajat Bhardwaj, Rishabh Bhardwaj, Soujanya Poria, Vinod SP","submitted_at":"2025-05-25T02:56:19Z","abstract_excerpt":"We assess whether AI systems can credibly evaluate investment risk appetite-a task that must be thoroughly validated before automation. Our analysis was conducted on proprietary systems (GPT, Claude, Gemini) and open-weight models (LLaMA, DeepSeek, Mistral), using carefully curated user profiles that reflect real users with varying attributes such as country and gender. As a result, the models exhibit significant variance in score distributions when user attributes-such as country or gender-that should not influence risk computation are changed. For example, GPT-4o assigns higher risk scores t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18953","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/2505.18953/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":"2505.18953","created_at":"2026-07-05T11:32:29.718619+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18953v2","created_at":"2026-07-05T11:32:29.718619+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18953","created_at":"2026-07-05T11:32:29.718619+00:00"},{"alias_kind":"pith_short_12","alias_value":"G4RM7W4NBCB3","created_at":"2026-07-05T11:32:29.718619+00:00"},{"alias_kind":"pith_short_16","alias_value":"G4RM7W4NBCB3PUJA","created_at":"2026-07-05T11:32:29.718619+00:00"},{"alias_kind":"pith_short_8","alias_value":"G4RM7W4N","created_at":"2026-07-05T11:32:29.718619+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27570","citing_title":"Auditing AI Investment Recommendations as Executable Actions","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4","json":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4.json","graph_json":"https://pith.science/api/pith-number/G4RM7W4NBCB3PUJAZN5TES6OP4/graph.json","events_json":"https://pith.science/api/pith-number/G4RM7W4NBCB3PUJAZN5TES6OP4/events.json","paper":"https://pith.science/paper/G4RM7W4N"},"agent_actions":{"view_html":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4","download_json":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4.json","view_paper":"https://pith.science/paper/G4RM7W4N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18953&json=true","fetch_graph":"https://pith.science/api/pith-number/G4RM7W4NBCB3PUJAZN5TES6OP4/graph.json","fetch_events":"https://pith.science/api/pith-number/G4RM7W4NBCB3PUJAZN5TES6OP4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4/action/storage_attestation","attest_author":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4/action/author_attestation","sign_citation":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4/action/citation_signature","submit_replication":"https://pith.science/pith/G4RM7W4NBCB3PUJAZN5TES6OP4/action/replication_record"}},"created_at":"2026-07-05T11:32:29.718619+00:00","updated_at":"2026-07-05T11:32:29.718619+00:00"}