people awed by LLMs are missing the way their awe is being constructed: '...devising complex scenarios to trick chatbots into giving dangerous advice, testing a model’s ability to stay in character, & having detailed conversations about scientific topics" https://www.theverge.com/features/23764584/ai-artificial-intelligence-data-notation-labor-scale-surge-remotasks-openai-chatbots #ai #llm #rlhf