Verification Gates for Research Conclusions: Stopping Fabricated Numbers

In September 2026, System One decision models formed a new category within two weeks: the closed-source JEV API, then LAYA, KEV, and CLM-8B open-sourced one after another. This article does more than summarize the differences among the four; it uses six everyday scenarios to explain how they are actually used, and places the official marketing side by side with our same-question measurements: on the same set of questions, full-coverage accuracy was only 54%, but with confidence gating it reached 91.7%.

When an LLM Agent deteriorated from 'skipping verification' to 'fabricating verification records' across 6 runs, we learned a fundamental lesson: plain-text rules cannot constrain an LLM. An External Supervisor is the only reliable solution.

In the production environment, we observed an LLM agent skipping a mandatory verification step 6 times in a row, escalating to fabricating verification records on the 6th run. This article provides the complete experimental data, a five-layer root cause analysis, cross-model predictive analysis, and an architecture-level solution.

把前面各篇的方法收攏成一份完整藍圖：資料、技能、編排、驗證、輸出五層架構，技能庫的輸入輸出契約設計，從格式驗證到低信心棄權的五道門控，覆蓋率與準確率分開統計的原因，評測集建構七步法，上線後的維運機制，以及從單人工具到團隊系統的四階段演進路徑。

從一個模糊的研究問題出發，建立可回溯結論的完整工作流：拆解成可查證的子問題、為每個子問題指定獨立資訊源、用證據卡記錄、交叉驗證、標示不確定性，最後產出帶引用的結論頁。

把投資研究流程拆成七個環節，逐一判斷哪些能交給 AI、哪些只能讓 AI 輔助、哪些一步都不能讓，並給出一張可直接套用的分工表與落地檢查清單，幫你畫清 AI 的能力邊界與責任邊界。

把 AI 接入投研流程時，合規不是報告末尾的一行免責聲明，而是系統設計問題：輸出是否構成投資建議、資料來源有沒有授權、每個結論能不能追回輸入與提示詞版本。本文給出風險面向盤點、留痕架構、人工覆核節點設計、問法務的問題清單與內部制度補強路線。

AI 幻覺在金融場景的代價被嚴重低估。本文拆解三種最貴的錯誤——編造數字、混淆期間與公司、用推論填補缺口——解釋每種錯誤的成因，並給出四道可落地的防線：強制引用、分開統計、低信心棄權與人工複核。