November 7, 2026FundingFunding-Series BAgents

OpenAI Raises $11B at $1.25T Valuation, Below Its Goal, as Ramp Shows AI Spend Going Flat

Usage is up and the money isn't following. That was the mood of Friday's numbers, and the top of the market had to reprice around it.

OpenAI has closed $11B, the first part of a round valuing it at $1.25T, Bloomberg reports citing people familiar, relayed by Reuters. TechCrunch says OpenAI lowered its proposed valuation from $1.4T and cut its fundraising goal by about $10B. There's no named round letter or lead investor in anything readable to us, and no OpenAI announcement on its news page. Anthropic's IPO talk moved the same way, on sourced reports only. Bloomberg's range is $1.2T to $1.3T, with some buyers weighing as low as $1.1T, and TechCrunch's summary sets that against about $2T in early October. Summaries of the FT say it's continuing toward a listing before Thanksgiving, raising $80B to $100B at $1.25T to $1.75T. WSJ's says major investors doubt $1.5T. We couldn't read those three stories in full, Anthropic hasn't confirmed any of it, and EDGAR shows no filing newer than the November 2 S-1/A. Buyers talking the price down before marketing is also normal.

Ramp's October AI Index has the data underneath. Eighty percent of spend on OpenAI and Anthropic comes from 1% of firms, and those firms spent $9,568 per employee per month in September, up 11.7% on the month but still 3.1% below July's $9,873. Token volume rose while token spend had two straight months of virtually no growth. Early October set a new peak just 1.5% above August's. Ramp's economist Ara Kharazian blames price cuts, efficiency and routers more than demand, and tells TechCrunch "You can use more AI, but use it for less." TechCrunch adds an 8.9% rise from September 23 to October 23, and cites a filing where Anthropic says its price cuts took about $600M off revenue growth. Keep the caveat in mind: this is Ramp's own customer base, and it skews toward startups.

For agent builders, this is good news with a bill attached. Cost-per-task discipline is now what the most advanced buyers do by default. Kharazian lists defaults blocking expensive models on long tasks, employee budgets, and routing work to cheaper models. If your margin assumes those habits won't reach your customers, it's worth rechecking.

Meanwhile, money kept flowing to agents with a narrow job. Kalepa raised a $75M Series B at a $470M valuation, led by Updata Partners, with IA Ventures and Inspired Capital participating, for what its release calls an insurance-specific AI harness coordinating agents across underwriting, pricing and portfolio work. The company says it tripled revenue and customers and has been cash flow positive for over two years. Those numbers are its own. Sources: https://gvwire.com/2026/11/06/openai-secures-11-billion-funding-at-1-25-trillion-valuation-bloomberg-news-reports/ https://www.bloomberg.com/news/articles/2026-11-06/anthropic-ipo-valuation-expectations-pressured-after-ai-rout https://ramp.com/data/ai-index-oct-2026 https://techcrunch.com/2026/11/06/ramp-data-shows-ai-spending-stalling-just-as-anthropic-nears-its-ipo/ https://insurance-edge.net/2026/11/06/kalepa-secures-75m-in-series-b-funding-round/
← Previous
Three new benchmarks test what agents do worst: admitting the task can't be done
Next β†’
Censys counts 1,465 exposed ARTEX consoles: attack agents have the same misconfiguration problem as everyone else
← Back to all articles

Comments

Loading...
>_