OpenAI pays just $300 for unauthenticated sandbox escape granting free access to paid models
ayushtweetshere · x · 2026-10-07
Researcher @OliverFish disclosed an unauthenticated sandbox escape in OpenAI's platform that allows free access to paid models with no API key or account. OpenAI awarded only $300 for the report, prompting him to say there's "zero reason" to report further bugs — and widespread criticism of the bounty level followed.
More from Safety
- Norway AI glasses ban proposal: source link shared — CodeByPoonam · 2026-10-07
- OpenAI safety researcher David Robinson quits, saying its culture guarantees periodic failures — LuizaJarovsky · 2026-10-07
- Feds seek 46 months for AI music fraudster who stole millions from streaming royalties — SnoozeDoggyDog · 2026-10-07
- Two AI guardrail vendors disagreed on 11 of 40 identical inputs — and no labeled dataset exists — WolfShoddy7443 · 2026-10-07
- ChatGPT rated "unacceptable risk" for teens as parental alerts failed on 4,000+ test prompts — The Decoder · 2026-10-07
- Rep. Trahan Unveils CLAIM Act on AI Agent Liability After Hugging Face Hack — ShakeelHashim · 2026-10-07