谁会想到在将人工智能部署到生产环境之前，还得先阅读其代码？我一直在强调和解释这一点：低级逻辑模型（LLM）并非确定性程序；它们可能会出现幻觉，而仅仅一次幻觉就可能毁掉你的业务。
本文为机器翻译
展示原文
Guillermo Rauch
@rauchg
03-04
A Vercel user reported an issue that sounded extremely scary. An unknown GitHub OSS codebase being deployed to their team.
We, of course, took the report extremely seriously and began an investigation. Security and infra engineering engaged.
Turns out Opus 4.6 *hallucinated a
来自推特
免责声明：以上内容仅为作者观点，不代表Followin的任何立场，不构成与Followin相关的任何投资建议。
喜欢
收藏
评论
分享
