字节跳动等机构发布75页论文，将AI自优化分为5级，当前多数模型处于低级阶段
"Everything AI has achieved so far is but a drop in the ocean." That's page one of a 75-page paper ...
朋友，字节跳动和清华等团队搞了个75页的论文，把AI自己优化升级的能力分了五级，现在看主流模型都还处在比较初级的阶段，挺有意思的。
这篇由字节跳动、清华大学等机构联合发布的75页论文，将AI自优化能力分为五个等级。在Level 1阶段，人类仍决定AI的改进方向；而Level 5则意味着AI能自主重写构建其后续版本的过程。论文对491篇现有论文进行分类，结果显示75%的模型处于Level 1或2，仅有不到6%达到Level 5。OpenAI、Anthropic和Google DeepMind等西方实验室已将AI自动化研发列为安全框架中的风险。
"Everything AI has achieved so far is but a drop in the ocean." That's page one of a 75-page paper ...
"Everything AI has achieved so far is but a drop in the ocean." That's page one of a 75-page paper called "The Last AI Built by Humans," from 30+ researchers at ByteDance, Tsinghua, Shanghai AI Lab and six other institutions. The paper ranks AI self-improvement on five levels. At Level 1, humans still decide what gets fixed. By Level 5, the AI is rewriting the process that builds its own successor. They also sorted 491 existing papers onto the ladder, and 75% landed at Level 1 or 2. Under 6% made it to Level 5. Level 5 is what Western labs have been warning about. OpenAI, Anthropic, and Google DeepMind each list AI automating its own R&D as a risk in their safety frameworks. 💬 3 🔄 0 ❤️ 5 👀 1308 📊 4 ⚡