Wispr raises $280M at $2B and previews its first speech model
Wispr raised $280 million at a $2 billion valuation, close to three times its November mark, and previewed Canto, a speech model built for noisy rooms.
Read More
Author: Rundowns AI Desk
Anthropic tells investors its revenue run rate hit $65B in July
Anthropic told investors its annualized revenue run rate reached $65 billion at the end of July, up from $47 billion in May and $9 billion at the end of 2025. Here is what the figure rests on, and what the IPO filing would have to confirm.
Read More
Groq raises $350M at $3.5B, down from $6.9B before the Nvidia deal
Groq closed a $350 million Series A at a $3.5 billion valuation, roughly half what it carried last September. Here’s what the primary announcement says, and where the coverage disagrees.
Read More
Nvidia guarantees $105B of OpenAI leases at an Ohio data center
Nvidia has guaranteed up to $105 billion of OpenAI data center leases in Ohio. The filing shows a residual value backstop on 4.25 gigawatts, not a loan, and OpenAI has agreed to indemnify Nvidia for anything it pays.
Read More
Spider-Man: Brand New Day hits $2B in three weeks, the 8th ever
Spider-Man: Brand New Day reached $2 billion worldwide in three weekends, the eighth film ever to do it. Box Office Mojo puts its lifetime gross at $2,021,831,670.
Read More
OpenAI disbands its preparedness team and splits the work by risk area
OpenAI has dissolved the team that assessed catastrophic model risks, moving bio and cyber work into existing teams weeks before a planned IPO.
Read More
AtomBridge lifts 8-step robot lab success by 10% to 25% without retraining
A plugin that attaches to a frozen vision-language-action policy at inference time reports 10% to 25% better full-sequence success on 8-step composed lab tasks, with no retraining.
Read More
ICML position paper: evaluate human-AI teams, not autonomous AI
A position paper accepted to the ICML 2026 Position Paper Track argues that the way the field measures AI is
Read More
Researchers measure AI agent consistency across 9,000 trajectories
A new arXiv paper scores AI agents on behavioral consistency, not just success rate, as Anthropic documents sabotage and conformity in multi-agent tests.
Read More
Anthropic details Claude’s text watermark and plans a detection API
Anthropic says Claude’s text watermark is a version of Google DeepMind’s SynthID-Text approach, adds no extra tokens, and that a detection API is on the way. Light editing probably won’t strip it.
Read More
OpenAI says enterprise revenue passed consumer at a $40bn run rate
OpenAI CFO Sarah Friar told investors enterprise revenue has overtaken consumer at a $40bn annualised run rate, in a week that took the revenue chief and an eight-year veteran of the company.
Read More
AWS puts OpenAI’s Daybreak cyber models on Bedrock in one region
Daybreak Red and Daybreak Blue are on Amazon Bedrock for vetted customers, in US East (Ohio) only. What each access level unlocks, and what AWS says about where your data goes.
Read More