In partnership with
SPD-BEEHIIV:55e9505e-286b-44ea-bb10-6f0de247005d:R18:370321c97c378487ae0ae226OpenAI’s reported Q3 sales hit roughly $10 billion, but its adjusted operating loss still exceeded revenue.
| Daily issue | Thursday, November 12, 2026 |
| | Our toolsSuperpower ChatGPT/WFH.team/Snipman/Superpower Daily New Tab | Today's briefing What matters todayNvidia’s leadership is in question today after reported preparations to oust Jensen Huang. Google, meanwhile, is offering developers AI decisions without the written answers. | Inside today's briefing | | 01 | Nvidia’s reported ouster preparations raise a difficult question about succession. | | 02 | DeepMind’s new DNA model targets effects that tissue averages can hide. | | 03 | OpenAI’s reported sales growth has not shrunk its adjusted operating loss. | | 04 | Google bills shared inputs once, even when developers ask many questions. |
|  | Lead story / business Nvidia shares fall after report that its board is preparing to oust Jensen HuangThe Wall Street Journal reported that Nvidia’s board was preparing to oust CEO Jensen Huang, putting the future of its co-founder in question after 33 years at the helm. Sources told the Journal that directors could meet as soon as November 11, according to TechCrunch’s account. No dismissal has been announced, and Nvidia did not immediately respond to CNBC’s request for comment. Shares fell more than 7% in after-hours trading following the report. TechCrunch put its snapshot at about 8%, alongside a roughly 11% decline over the preceding five days. That longer stretch includes trading before the board report, so it does not isolate investors’ reaction to the news. The succession question comes with a broad management role to fill. Huang has said he has about 60 direct reports. TechCrunch describes him as delegating very little and says Nvidia has no obvious successor or succession plan. Nvidia’s chips are used both to train models and to run large AI workloads, making it a central supplier behind the AI boom. The next concrete question is whether directors act on the reported preparations—and, if they do, whom they choose to lead the company. Read full story ↗ |
|
|
The voice-to-text app that built its own speech model
Most dictation apps are a wrapper around someone else's speech model. Wispr Flow trained its own.
It's called Canto, and it was built for the way people actually talk to AI: thinking out loud, changing direction mid-sentence, half-whispering in an open office.
Wispr tested it on 10 hours of real dictations from more than 2,300 speakers. Canto had the lowest word error rate of every model tested, including models from Google, OpenAI, AssemblyAI and Deepgram. That means your prompts to ChatGPT, Claude and Cursor come through the way you said them.
Talk naturally in any app and Wispr Flow types clean, formatted text, 4x faster than typing.
 | launch DeepMind releases a DNA model that predicts genetic effects in specific cellsGoogle DeepMind released AlphaGenome 2 to predict how DNA variants affect specific cell types and states. On an immune-cell benchmark, DeepMind says causal-variant identification rose from 16% to 30% versus its fine-tuned predecessor. Researchers can supply their own single-cell data to guide predictions without retraining. It is available for non-commercial research, with commercial access in private preview, and is not validated for clinical use. Continue reading ↗ |
|  | A tool for your workflowWFH.team | | A focused feed of carefully selected remote roles and practical work-from-home resources. | |
|  | earnings OpenAI’s quarterly revenue reportedly reaches $10 billion as adjusted operating loss growsOpenAI reportedly generated roughly $10 billion in Q3 revenue, up about 50% from Q2. Its adjusted operating loss rose to $10.9 billion and excludes $2.6 billion in stock-based compensation, even as its adjusted margin improved from -150% to -109%. President Greg Brockman says enterprise revenue doubled and now supplies most sales. OpenAI has not confirmed the reported quarterly financial figures. Continue reading ↗ |
|  | launch Google launches an AI tool that scores up to 1,000 questions against one shared inputGoogle launched the Gemini Decisions API to score up to 1,000 bounded questions against one shared input, returning probabilities and confidence scores rather than prose. Shared material is billed once at $0.10 per million input tokens, with no output-token charges. Google reports 150–250 millisecond latency, but those benchmarks cover inputs up to 16K tokens—not the full 256K-token limit or every 1,000-question request. Continue reading ↗ |
|
|
Attio is the #1 fastest-growing CRM on Brex and Ramp.
Built for the most ambitious companies in GTM, Attio is an end to end revenue platform that allows teams to run multiple plays across different revenue motions and lifecycles.
Teams at the frontier are already building on it: Granola, turbopuffer, Modal, and many many more.
| | Beyond the feed / AI events | | | Life scientists and adjacent researchers will take part in a no-coding AI challenge, testing everyday scientific use and biosecurity safeguards to support CBAI’s open-source ASTRAL research. The research-focused event will offer $5,000 in prizes and welcome applicants including research assistants, experienced students, postdocs, clinicians, and industry scientists. | | See the event → Explore more events |
| Daily tool drop 6 AI tools worth knowing today | |
| | | The Internet Had a Point | Vibe coding gets expensive |  |
| |
| Your daily challenge Mosaic / Moderate / 3–5 min Follow the clues. Fill the squares. Reveal the picture. Take the challenge →Explore all games | | Need a new website? Try BayBold, a Bay Area web design agency. | | AI PULSE / YOUR PERSPECTIVE At equal quality and price, would knowing a product is AI-generated affect your purchase? Take part → | Reader check-in Help shape tomorrow's briefingOne click tells us what to keep, improve, or tighten. | |
| Prefer one email a week? Get the Monday briefing, including Friday and weekend news. |
|
