Show and guest

Yahoo Finance interview with OpenAI CFO Sarah Friar. Original English title: OpenAI's Comeback, Explained by Its CFO.

Key takeaways

  • Q3 was a “blockbuster quarter”: annualized revenue run rate approached 50 billion dollars, up 70% quarter over quarter, driven by GPT-5.6 at the start of the quarter and Astro, launched in early September 0:03
  • The enterprise business doubled in Q3, with revenue split evenly between enterprise and consumer; coding, cybersecurity, and virtual assistant Dots are entry points into businesses. Visa’s first deal was in cybersecurity, while Amgen values IP security 0:21
  • On the consumer side, there are 1.2 billion weekly users, up 20% quarter over quarter; 300 million people ask health questions each week, and 200 million ask about personal finance 1:27 4:22
  • In Q3, paid subscribers reached 100 million, while 90% remain free. Dots will use subscriptions in the short term, with ads and e-commerce possible over the longer term; starting with enterprise is “high-altitude training,” and it will be rolled out to consumers before year-end 6:08
  • The cost of the Luna model is about 1/5–1/6 that of Chinese open-source models. After prices were cut by 80%, usage grew nearly 10-fold. She says the idea that “Chinese models are cheap and almost as good” is a myth that must be dispelled in 2026 7:06 8:04
  • On sovereign models, she cites Stanford data: in 2025, the United States had 52 frontier models, China 37, and Europe only 1. She believes that countries wanting their own AI now is not a wise outcome; the key is to be a trusted partner 10:00

Background and notes

Yahoo Finance interviewed OpenAI CFO Sarah Friar about Q3 results, the enterprise business, Dots, and model costs. Subtitle coverage: the video’s subtitles contain 330 timestamped segments, covering 0:00–11:25 (total length 11:23), and cover the entire video. The views and timestamps in this article are based on the original subtitles; each timestamp marks the start of the corresponding spoken segment. Publication-time note: the publication time was 2026-11-11 Hong Kong time, at approximately 11:00, converted from the relative time displayed by YouTube when retrieved, with hour-level precision.

Source links

Further reading

Related topics