we wouldn’t need CoT monitoring if AI was trustworthy enough to follow instructions. but it isn’t. ...
Gary Marcus(@GaryMarcus) · 人工智能
we wouldn’t need CoT monitoring if AI was trustworthy enough to follow instructions. but it isn’t. and that—not monitoring—is really the core problem. 💬 0 🔄 2 ❤️ 11 👀 1280 📊 2 ⚡ Powered by xgo.ing