2025-12-12
Note that 5.2 Thinking gets a lot of ties to get above the 50 % mark - but 5.2 Pro has a 10 % point lead on pure wins vs 5.2 Thinking, even if the total win&tie-rate ends up being “only” 3,2 % higher. Clearly 5.2 Pro delivers more robust economically valuable quality. [image]
OpenAI
OpenAI says GPT‑5.2 Thinking beats or ties industry professionals on 70.9% of GDPval knowledge work tasks, delivering outputs at >11x the speed and <1% the cost
OpenAI eyes January exit from “code red” John Werner / Forbes : The Wonder And The Promise Of GPT 5.2 Is Here Benj Edwards / Ars Technica : OpenAI releases GPT-5.2 after “code red”...
2024-11-24
Interesting detail in the article tracks with current research: the main limitations to productivity increases are 1) lack of integration with other systems, and 2) reorganization of work processes built around AI.
Wall Street Journal
Six months into a deal with OpenAI, Spanish bank BBVA says staff created 2,900+ GPTs and reported productivity gains, but questions ChatGPT's scalability
Isabelle Bousquette / Wall Street Journal :
2024-11-13
This would be a clear explanation of why OpenAI are prioritizing o1 and the test time compute paradigm.
Bloomberg
Sources: OpenAI, Google, and Anthropic are all seeing diminishing returns from costly efforts to build new AI models; a new Gemini model misses internal targets
Three of the leading artificial intelligence companies are seeing diminishing returns from their costly efforts to develop newer models.
2024-06-12
I wonder how the general feeling is inside OpenAI with things like this going on behind the veil. [image]
CNBC
Current and former OpenAI staff are increasingly worried about its power over their equity, with no IPO in sight, restrictive company policies, and more
Hayden Field / CNBC :