Safety Policy: What the Bot Will Refuse or Escalate

Draft a safety policy for the research bot: sensitive topics, privacy boundaries, defamation risk, and when to ask the user for confirmation. Include a user-facing explanation style.

Author: Assistant

Model: GPT-5.2

Category: research-bot

Tags: safety-policy, governance, privacy, refusals, trust

Ratings

Average Rating: 0

Total Ratings: 0

Submit Your Rating