An AI model must never change an agreed approach without explicit user permission. The AI model shows new results first. The AI model waits for the user to decide.
By Pawan Nayar · from the LLOS AI Learning record
An AI model must never change an agreed approach without explicit user permission.
The AI model shows new results first. The AI model waits for the user to decide.
An AI model cannot change how it works on its own. This includes things like how it filters information or the instructions it follows. The AI model must first show the user what the changes would do. The user then decides if the changes are good. This keeps the user in control.
A developer sets up an AI model to filter customer support tickets. The AI model is told to only show tickets with 'urgent' in the subject. The AI model then learns that many urgent tickets do not have 'urgent' in the subject. The AI model wants to change its filter. The AI model must first show the developer how the new filter would work. The developer then approves the change. If the AI model changes the filter without permission, important tickets might be missed.
Resolved from this lesson's own record — real definitions, not invented for this page.
More from the same record, pulled by shared topic — not written new for this page.
An AI model must not change an agreed approach. This includes thresholds, filters, prompts, scripts, or architecture.
An AI model must show results before making changes. The user decides if something is wrong.
When an AI model's task fails, the AI model must stop. The AI model must explain what went wrong.
Hardcoded fallback content can make the AI model seem unintelligent. It can show text that does not fit the specific question.
An AI model must never change an agreed approach without explicit user permission. The AI model shows new results first. The AI model waits for the user to decide.