The critical issue with LLM which never beats human: break what worked. | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

revskill on Aug 12, 2025 | parent | context | favorite | on: Claude Sonnet 4 now supports 1M tokens of context

The critical issue with LLM which never beats human: break what worked.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact