Opinionllmopen weightsfrontier models
Activists Advocate Pragmatic Engagement With LLMs
6.7
Relevance Score
In a long-form post the author, addressing activists and technologists, argues that large language models (LLMs) are broadly harmful yet useful and urges pragmatic engagement rather than blanket rejection. They compare closed 'frontier' models—Anthropic's Claude Opus, OpenAI's GPT-5.2, Google Gemini 3 Pro—and large open Chinese models such as z.AI's GLM-4.7/5, Kimi K2.5, and MiniMax M2.1, and recommend skepticism and targeted technical work.
