OpinionSimon Willison

2026 in LLMs (so far)

#llms#ai#coding agents#software engineering#security

English

In 2026, significant advancements in large language models (LLMs) were marked by the release of Claude Opus 4.5 and GPT-5.1, which improved coding agents' reliability. The year also saw a growing focus on agent security and the implications of AI on software engineering, leading to discussions about the rapid changes in the profession.

中文

2026年，大型语言模型（LLMs）取得了重要进展，发布了Claude Opus 4.5和GPT-5.1，提升了编码代理的可靠性。今年还越来越关注代理安全和人工智能对软件工程的影响，引发了对这一职业快速变化的讨论。