# Khosla Ventures is in talks to raise $5.5 billion as Vinod Khosla doubles down on AI > Source: > Published: 2026-07-23 02:06:44+00:00 *Bloomberg reports Khosla Ventures is seeking $5.5 billion across new funds, nearly double what it sought last year and a clear signal that the most AI-committed firms in venture are still pressing harder.* The number tells the story. In February 2025, The Wall Street Journal reported that Khosla Ventures was seeking $3.5 billion across three funds. In 2023, TechCrunch reported that the firm was going after about $3 billion. Now Bloomberg says the Menlo Park firm is in talks to raise $5.5 billion across a new set of funds. That's a sharp jump at a time when many limited partners are still telling venture managers to slow down, prove returns, and come back with cleaner marks. Vinod Khosla doesn't appear to be in the mood for patience. Forbes says Khosla Ventures put $50 million into OpenAI in 2019 at a roughly $1 billion valuation, when the company still had little commercial infrastructure and no mass-market product. Forbes also reported that OpenAI was valued at $852 billion as of March 2026, after a $122 billion funding round, and a reconstructed cap table published by Forbes put Khosla Ventures' current OpenAI stake at about $1.5 billion. That kind of return doesn't just validate a thesis. It becomes the thesis. It is not subtle. Forbes ranked Khosla first on its 2026 Midas List, with OpenAI listed as his notable deal. When your firm was the first venture investor in one of the most valuable private companies in the world, raising a larger pool of capital isn't automatically hubris. It is the obvious move if you believe the next OpenAI-sized opportunity is still ahead of you. ## The arms race around AI funds The competitive context matters here. Andreessen Horowitz said in January that it had raised more than $15 billion across new funds, including $6.75 billion for growth, $1.7 billion for apps, $1.7 billion for infrastructure, $1.176 billion for American Dynamism, $700 million for Bio and Health, and $3 billion for other strategies. Sequoia, according to Bloomberg, raised about $7 billion in April for a new fund aimed at bigger bets, including AI companies with huge compute needs. Scale counts. In that environment, $5.5 billion is not a strange number. It is the price of staying in the room when foundation model labs, AI infrastructure companies, and defense startups can demand checks that would have looked absurd in venture a decade ago. Khosla's recent deal flow shows the same direction. Bloomberg Law reported on July 7 that legal AI startup Norm Ai raised $120 million at a $1.2 billion valuation in a round led by Khosla Ventures. On July 20, Twenty, an Arlington, Virginia cyber warfare startup founded in 2024, announced a further $30 million investment from Khosla Ventures at a $1.2 billion valuation, after a $100 million Accel-led Series B in June. One company is trying to automate legal work. The other is building AI-enabled systems for the U.S. military and intelligence community. Different markets, same belief: AI is not a feature layer. It is where whole categories get rebuilt. Founders know this. A larger fund gives Khosla Ventures staying power in a negotiation, especially when a company needs a large follow-on round and doesn't want an investor that will hesitate at the size of the check. That can change allocation, pricing, board dynamics, and the speed of a deal. None of that shows up neatly in a pitch deck. ## What founders should take from it The broader signal is that the AI investment cycle has not cooled at the top. Yes, weaker companies are going to struggle. Yes, many LPs are still tired of paper gains and delayed exits. But the best-positioned firms are not behaving as if the AI trade is over. Andreessen Horowitz said its January raise represented more than 18% of all venture capital dollars allocated in the United States in 2025. Khosla's reported $5.5 billion target fits that same pattern. Here's the thing: this is not free money for every startup with an AI slide. More capital at the largest firms means more competition for the best companies. But it also means the bar rises for everyone else. If you're raising a Series A in AI infrastructure, legal AI, cyber, robotics, or applied enterprise AI, you may find investors willing to move quickly when the company is real. If your pitch sounds like five other companies in the same week, the big funds won't save you. The bar rises too. Khosla told Bloomberg in April that OpenAI's $852 billion valuation was not too high. You can agree with that or not, and plenty of investors quietly don't. But his record on this specific kind of call is hard to ignore. The $5.5 billion target, if Bloomberg's report plays out, is Vinod Khosla saying the AI cycle still has room to run. Fundraising talks can change before a close. That is the remaining caveat. But the direction is clear enough: the venture firms that won the first phase of AI are trying to buy the right to compete in the next one. **Also read:** [Trump's push to ban Chinese open-weight AI would kill US startups while leaving the models freely downloadable](https://startupfortune.com/trumps-push-to-ban-chinese-open-weight-ai-would-kill-us-startups-while-leaving-the-models-freely-downloadable/) • [Cursor Router picks your AI model for you and cuts coding costs by 60%](https://startupfortune.com/cursor-router-picks-your-ai-model-for-you-and-cuts-coding-costs-by-60/) • [Wall Street coined a new $3 trillion sector called Memi and the stocks powering it are already in a bear market](https://startupfortune.com/wall-street-coined-a-new-3-trillion-sector-called-memi-and-the-stocks-powering-it-are-already-in-a-bear-market/)