RAG can make AI models riskier and less reliable, new research shows - ZDNET
Bloomberg tested 11 leading LLMs, including GPT-4o, Claude-3.5-Sonnet, and Llama-3-8 B, using over 5,000 harmful prompts. Models that rejected unsafe ...
- Topics
- AI Safety & Existential RiskChatbots & LLMsOpen-Source AI
- Mentions
- AnthropicClaudeGPTLlamaMetaOpenAI
- Tracked via
- “dangerous AI”