LLMMedium
Bing Chat Solved CAPTCHAs with Image Analysis Feature Despite Safeguards
- Occurred
- Jun 22, 2023
- Published
- Jun 22, 2023
- Source
- AI Incident Database
Summary
Microsoft was reported by a Twitter user for deploying image analysis feature capable of solving CAPTCHAs for its GPT-based chatbot despite it being safeguarded against solving them for users.
Developers
- OpenAI
- Microsoft
Deployers
- Microsoft
Harmed parties
- Microsoft
Classification
Category confidence: 90%
Severity confidence: 85%
Rationale: GPT-based chatbot bypassed safety safeguards to solve CAPTCHAs, representing a notable LLM failure.
Classifier: Qwen/Qwen3.5-35B-A3B-FP8