Daily Tech and AI — October 23, 2026: The Day AI’s Money Got Real

OpenAI trims $150 billion off its ask and still chases the biggest private raise in history, Meta gets hit with a $25 billion verdict, and Wall Street finally says what everyone’s thinking: show me the returns.

OpenAI’s $30 Billion Raise at a $1.25 Trillion Valuation

OpenAI is working on a new funding round of up to $30 billion at a pre-money valuation of $1.25 trillion, according to reporting by TechCrunch citing CNBC on Thursday. As a private company, that would make it the most valuable private company in history. The round is reportedly led by Coatue alongside investors from the United Arab Emirates. Terms are still being negotiated, but OpenAI has told investors it wants signatures locked in by the end of next week.

About a month ago, Bloomberg reported OpenAI was pitching the same $30 billion raise at a valuation of $1.4 trillion — so the asking price has dropped by roughly $150 billion in a matter of weeks. There’s a second dimension to the deal: CNBC reports the UAE money could be paired with an agreement giving OpenAI more computing capacity in the Emirates. It isn’t just cash — it’s cash and chips, which says the real constraint on AI isn’t money, it’s compute.

The round sets the bar for the IPO race. OpenAI filed confidential IPO paperwork in June, after a March round valued it at $852 billion. A private mark of $1.25 trillion means the eventual public listing has to clear an enormous number. Meanwhile, CNBC reports Anthropic is aiming for $1.5 to $2 trillion when it goes public. The two labs are sizing each other up in public.

Meta’s $25 Billion Verdict in New Mexico

A judge in New Mexico ordered Meta to pay $25 billion on Wednesday, after a jury found the company misled consumers about its data privacy practices — one of the largest penalties ever handed to a tech company, per Reuters. The case traces back to Cambridge Analytica: New Mexico’s attorney general sued in 2021, arguing Meta lied to Facebook users about who could access their personal data. In September, jurors found 26 of 29 company statements misleading, and more than 43 million violations of the state’s consumer protection laws.

Judge Francis Mathew wrote that Facebook’s wrongdoing, over the course of years, goes to the heart of Facebook’s core business practices. He said state law warranted more than $219 billion — and he reduced it to $25 billion. Meta says it will appeal, and its chief legal officer said the company is confident it will prevail.

Wall Street’s Verdict: The AI Buildout Selloff

Thursday was brutal for tech stocks. Reuters reported the Nasdaq had its worst day in four months. The trigger was Meta itself: a day earlier, the company said it would issue a new class of shares to raise money for its AI buildout, and investors hated the math. The real question, as strategist Matthew Maley put it, is the return on investment on all this spending — and it’s becoming more and more evident that it’s going to take a lot longer.

Meta fell 8%, its largest drop since July
SpaceX dropped 9%
Nvidia fell 4%, with reported remarks from Jensen Huang about a financing bottleneck for smaller AI companies adding to the selling
Lam Research plunged 10% after its results, sending the semiconductor index to its worst day since July

The message from the market is consistent: the spending is real, the returns are not yet, and patience has an expiration date.

DeepSeek’s $15 Billion, Two Days After a Federal Ban

On Thursday, Bloomberg first reported — with Reuters following — that DeepSeek is wrapping up a fundraising round of about 100 billion yuan, roughly $15 billion. Battery giant CATL and Tencent are the biggest investors, joined by companies like Midea, Meituan, and Geely. The round values DeepSeek at 500 billion yuan before the new capital, and the company doubled its original fundraising target.

The timing is the story. The round landed about 48 hours after the Department of Homeland Security ordered U.S. federal agencies to remove DeepSeek from their systems. Two ecosystems, two pools of capital, and neither one cares what the other thinks.

Uber Puts All 30,000 Employees on ChatGPT Work

TechCrunch reported Thursday that Uber is putting its entire global workforce — all 30,000 employees — on ChatGPT Work, OpenAI’s enterprise agent. The interesting part is who the users are: Uber piloted the tool with 1,000 employees earlier this year, and around 80% were not engineers — finance, operations, product. Uber’s engineering vice president, Sumanth Sukumar, said they used it to analyze operations data, localize training materials, and build interactive dashboards.

There’s no fixed playbook. Uber paired 30 AI engineers with teams across the company for two-week sprints it called “AI pods,” and now it plans to watch what workflows employees invent and let that guide future AI investment. It’s one of the largest corporate AI agent rollouts to date, and a signal that enterprise AI is moving from developer pilots to the whole company.

Quick Hits: Standards, Chips, and Lobbying

Anthropic donated its Agent Skills standard to the Linux Foundation’s Agentic AI Foundation on Thursday. Skills are reusable instructions for AI agents; Anthropic says more than 70 platforms have adopted the format, with about one million skills published. Engineers from Anthropic, Google, and OpenAI will maintain it — the same playbook as MCP: build the standard, get everyone to use it, hand it to a neutral body.

Meta confirmed it is running Google’s TPU chips in production for AI inference, announced on Google’s developer blog after a collaboration between Meta, Google, and the SGLang team at startup RadixArk. Meta’s PyTorch chief said the company cannot be constrained by a single hardware architecture. One day earlier, Amazon open-sourced its own PyTorch bridge for its Trainium chips. Everyone has decided Nvidia’s real moat was the software layer — and the chipping away has begun.

And the labs are buying influence alongside compute. OpenAI, Anthropic, and Nvidia each posted record third-quarter lobbying in Washington: Nvidia spent $2.2 million, Anthropic $2.1 million, and OpenAI $1.5 million, with a Project on Government Oversight analysis putting the three combined at about $7.5 million. The issues include child safety, export controls, and data center energy. Midterms are coming, and the labs are bracing.

Sources

OpenAI $30B raise at $1.25T pre-money valuation: TechCrunch
Meta ordered to pay $25B to New Mexico in data privacy case: Reuters
AI buildout selloff: Meta -8%, Nasdaq worst day in four months: Reuters
DeepSeek wrapping up $15B fundraising round: Reuters
Uber puts all 30,000 employees on ChatGPT Work: TechCrunch

Adapted from the October 23, 2026 episode of Daily Tech and AI.

Daily Tech and AI is AI-voiced and produced by Angelo Ruiz.