Create advanced Works with: Any AI

Handling Negative Public Comments About Safety or Quality

Structured brief

Respond carefully and escalate.

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it.

Rate this prompt

Loading ratings…

Prepare responses. Product: [PRODUCT]. A comment alleges: [ALLEGATION]. What we know: [WHAT_WE_KNOW].

Draft a public holding reply under 60 words, a private message under 80 words and an internal escalation note. Do not admit or deny what is not verified. Define who investigates and the time frame. Safety issues go to the proper team immediately and may need regulator notice [VERIFY].

Return the replies and a log template.

Add a worked response timeline: within one hour acknowledge publicly, within four hours the safety or quality team reviews, within one day a decision on the next public update. Provide a table with who is responsible, what evidence is gathered, and what is said at each stage. If the allegation involves health or injury, direct the person to appropriate services and escalate to the responsible manager immediately. Reporting duties to regulators may apply and need review [VERIFY]. Keep a log of every statement and its time.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.

Related prompts