Engineer Sophie Alpert mandates that developers stand behind every sentence produced by LLMs. This policy bans the excuse that AI wrote a specific line during peer reviews. It treats AI writing as a lossy transformation of intent. Practitioners must manually verify that all generated documentation accurately represents their own thoughts before sharing.