Here are today's news for 12th of November about OpenAI, Anthropic, Meta and Microsoft. Let's dive into the details. OpenAI told investors on a call Tuesday, Nov. 10, that revenue for the quarter ending Sept. 30 reached $10 billion for the first time, Bloomberg reported, citing people familiar with the figures. That was nearly 50% above the roughly $6.7 billion reported for the second quarter and almost three times revenue a year earlier; the company’s annualized revenue pace was about $50 billion. CFO Sarah Friar said the growth was driven by frontier research and enterprise demand: enterprise revenue doubled quarter over quarter, while consumer subscriptions exceeded 100 million users. She said revenue was split evenly between enterprise and consumer. The figures signal a growing market for paid AI services and intensify competition with rivals such as Anthropic, while giving customers and investors evidence of adoption beyond ChatGPT’s consumer audience. But growth has not yet meant profitability: reports put OpenAI’s adjusted operating loss at $10.9 billion in Q3, up from $10. On Nov. 10, 2026, Anthropic launched its IPO roadshow and disclosed a proposed price of $70, $78 for 825 million Class A shares, seeking up to $64.35 billion before expenses. The sale comprises about 801.35 million newly issued shares and 23.65 million from existing holders; underwriters may buy another 123.75 million. At the top of the range, the Claude maker’s fully diluted valuation would reach roughly $1.4 trillion, far below recent $2 trillion speculation. Anthropic has applied to list on Nasdaq as ANT, with trading expected Nov. 19 and pricing targeted for Nov. 18. The prospectus also describes a separate $26 billion private placement involving strategic investors including Google and Nvidia. The scale makes the offering a test of investor appetite for AI companies amid market jitters and scrutiny of the sector’s enormous funding needs, while potentially providing a public, market benchmark against rival AI developers. On November 9, 2026, Meta CEO Mark Zuckerberg announced that Muse Ember, Muse Spark 1.3 and Muse Wonder were available as managed APIs through Google Cloud’s Model Garden. The three models target different workloads: Ember handles efficient, repeated tasks; Spark 1.3 supports coding and agent workflows, including tool use; and Wonder is aimed at complex software engineering. Google CEO Sundar Pichai welcomed the addition, saying Google was “excited to serve our Google Cloud customers.” For developers and businesses already using Google Cloud, managed access removes the need to host the models themselves and puts Meta’s offerings alongside Google’s models in one cloud environment. That broadens customer choice and gives Meta a route to enterprise users beyond its own services, while intensifying competition among cloud platforms to offer models from multiple providers. Google’s Model Garden describes the Muse lineup as multimodal; its models support context windows of up to 1 million tokens. On July 24, Nvidia, Microsoft, Meta and more than 20 other technology companies and organizations issued “Open Weights and American AI Leadership,” urging U.S. policymakers to avoid premature restrictions on open, weight AI. The letter argues that downloadable, modifiable models widen access for startups, businesses, researchers and public institutions, while broad limits could stifle competition and send development overseas. It warns that closed systems are not automatically safer: they “can be breached, misused, or fail in ways that outsiders cannot detect,” and says concentrating advanced capabilities in a few closed models compounds the risk. The signatories acknowledge concerns about unlawful extraction of value from proprietary models, but call for “targeted legal and commercial frameworks” rather than sweeping restrictions, including on distillation. They also urge policymakers to expand compute access and invest in shared datasets, tools and evaluation frameworks. OpenAI released GPT, 6 Cyber on November 5, 2026, making it available initially to a small group of vetted Daybreak Red customers through its API and Codex. The company says researchers used it to identify 81 previously unknown Firefox vulnerabilities, all rated high severity by Mozilla, and to reproduce 1,724 Linux kernel flaws and build working exploits for 948. Of those Linux findings, 588 were previously unknown; OpenAI said it reported 304 critical issues to maintainers with proposed patches. Mozilla has begun issuing fixes. The results matter because browsers and Linux underpin widely used consumer and server systems, while the model can both locate flaws and demonstrate how they may be exploited. OpenAI rated it “Critical” for cybersecurity under its Preparedness Framework, up from “High” for GPT, 5.6 Cyber. On ExploitGym, it scored 48.5%, versus 32.1% for its predecessor. That's all for today's AI news briefing. Have a great day!