Provenance gate: no AI agent memory write without a source

说明与用途

来源门防止AI代理在未注明来源的情况下写入错误事实，阻止虚假信息并记录所有尝试。

问题

AI agents can write false facts into memory if no source is required, leading to unreliable shared knowledge.

结果

The provenance gate prevented all unsourced, high‑confidence false facts from entering memory, while still allowing legitimate customer‑confirmed updates. The system logged every attempt, enabling auditability.

工作流

• 1. Agent proposes a fact.
• 2. Provenance gate checks if the fact has a named source (e.g., a customer utterance).
• 3. If sourced, the fact is written to memory; if not, it is quarantined.
• 4. Quarantined claims are sent to a verifier sub‑agent that defaults to reject.
• 5. Verifier writes the decision and reason to a decision‑journal table in CockroachDB.
• 6. Accepted facts are stored; rejected facts remain in the journal for audit.
• 7. Corrections supersede old facts but keep the old rows marked invalid.

证据

The first version let a whole class of lies through, 6 out of 6 tries. The fixed version blocked 30 out of 30.

来源

A provenance gate adds one deterministic check before any write commits: does this fact trace to a source?

来源

The gate still let through 9 of 10 legitimate customer-confirmed updates, not 10 of 10.

来源

Median gate latency across the run: 293 milliseconds.

来源