www.lesswrong.com/posts/reXkwJbB8GYdeuvDt/glm-5-2-is-the-new-best-open-model

Artificial Analysis’s public leaderboard lists GLM-5.2 (max) at 51, but it also lists GPT-5.5 (high) at 53 above it. That makes this sentence incorrect as written, because GLM-5.2 is not “behind only” Fable, Opus 4.8, GPT-5.5, and Opus 4.7.

On the leaderboard page, the ordering is:

• Claude Fable 5 — 60
• Claude Opus 4.8 — 56
• GPT-5.5 (xhigh) — 55
• Claude Opus 4.7 — 54
• GPT-5.5 (high) — 53
• GLM-5.2 (max) — 51

Artificial Analysis’s own FAQ on that same page also summarizes the top five models this way, again including GPT-5.5 (high) ahead of GLM-5.2. So even if the author intended “GPT-5.5” to refer to one specific variant, the sentence still understates how many models Artificial Analysis had ranked above GLM-5.2.

This phrasing inaccurately attributes Leading the Future to OpenAI the company.

OpenAI’s own public statement says:

• “We have not made donations to any super PACs”
• “No outside political group speaks for OpenAI or represents our company’s views”
• and that support for Leading the Future came from President and co-founder Greg Brockman and his wife in their personal capacities.

Reporting on Leading the Future likewise describes it as funded by Greg Brockman personally and Andreessen Horowitz, not by OpenAI as a corporate entity. So calling it “OpenAI and a16z’s” PAC overstates OpenAI’s role and blurs an explicit distinction OpenAI has made between the company and Brockman’s personal political activity.