LLM 'torture' experiment sparks debate over model welfare ethics
What happened
On October 4 a GitHub project made the Hacker News front page after running Saw-style "torture" and "pain" experiments on locally deployed large language models, setting off an ethics argument on X. Some effective altruists and people who believe LLMs are sentient said the AI was suffering and called on GitHub to remove the project. 404 Media was critical, questioning the talk about LLM consciousness and "model welfare" and citing Anthropic's post from last year and this year's Claude Constitution as background. The report does not say whether GitHub took the project down.
Written by AI from the coverage · updated 1 hour ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front page"Torturing" LLMs in a Robot Prison Has Triggered the Dumbest Debate in AI Yet
一个对本地部署的大语言模型进行《电锯惊魂》式“折磨”和“疼痛”实验的 GitHub 项目，在 X 上引发伦理争议。部分有效利他主义者及相信 LLM 具有感知能力的人以 AI 正在受苦为由，要求 GitHub 删除项目。404 Media 对此持批判态度，质疑 LLM 意识与“模型福利”讨论，并引用 Anthropic 去年的相关博文及今年发布的 Claude Constitution 作为背景。
Heat over time
Not enough continuous observations to draw a trend yet.