Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult

6 points by jonesn11 175 days ago | 1 comment

musha68k 175 days ago |

Prompt injections + context window engineering are the combined Archilles heel of the "agentic revolution".

musha68k 175 days ago |

Prompt injections + context window engineering are the combined Archilles heel of the "agentic revolution".