Trust and Safety

LLM Abuse Detection: What OpenAI's Scam Ban Reveals

LLM abuse detection failed at the message level and worked at the account level. OpenAI's Cambodia scam ban shows which signal actually catches misuse.