精选理由
Boris Cherny说多叠几层防御，间接提示注入能降到接近零，下周Claude Code的Auto Mode也默认开了，搞安全的可以看看。
Boris Cherny（@bcherny）在推文中称，通过叠加模型训练、输入探测和意图检查分类器等多层防御，可将间接提示注入在未见攻击上的成功率降至接近0。他表示一年前没预料到这一结果。Cherny同时宣布，Claude Code的Auto Mode将于下周起默认启用，细节见claude.com/blog/auto-mode。
原文 · AI Will
源：https://t.co/2SKaaqzZ4Y
源： x.com/bcherny/status… Boris Cherny @bcherny turns out you can get indirect prompt injection to ~0 on unseen attacks if you stack enough layers (model training + input probes + a classifier checking intent). didn't expect that a year ago. auto mode is default in claude code as of next week claude.com/blog/auto-mode… 🔗 View Quoted Tweet 💬 0 🔄 0 ❤️ 1 👀 438 ⚡