Has AI broken the old VC model?

Venture capital has always been a feast-or-famine industry, where most of the gains are concentrated in a small number of investments and a handful of funds. But amid the spate of mega-initial public offerings triggered by the AI boom, this is being stretched to an extreme. The question is whether this is just the latest manifestation of an old dynamic, or whether it will break the old way of doing business altogether.

It has never been more important to back the winners. Anthropic’s investors are expecting the company to reach a valuation of $2tn when it goes public in the coming weeks. Add in SpaceX, which began trading at $2tn after its IPO in June, and OpenAI, which is considering raising money privately at $1.2tn ahead of a public listing next year, and these companies alone could be worth well north of $5tn.

Now compare that with the entire history of IPOs from 1980 to 2025. The 3,365 tech companies that went public in that period were worth a combined $4.1tn when they started trading, according to data compiled by Jay Ritter, emeritus professor at the University of Florida’s Warrington College of Business.

The figure would be larger if adjusted for inflation, but even this comparison gives some sense of the enormity of the moment.

What makes this all the more striking is that it comes at a time when the VC industry is suffering a severe case of indigestion. When US interest rates started to rise in November 2021, it left a generation of unicorns (private tech companies valued at more than $1bn) high and dry. With the stock market no longer willing to pay up for high-growth tech beyond a narrower group of AI names, few have wanted to bite the bullet and go public at a lower figure.

The value of all the private unicorns, based on their latest funding rounds, has reached an astounding $5.3tn, according to PitchBook. While the Anthropic and OpenAI IPOs should eat into the overhang, it is an open question how much of the rest will ever see its way into investors’ pockets. As Ritter points out, the opaque way in which venture funds value their holdings gives them ample room to show investments that have declined at their historic value, while ones that have done well are marked to market.

Venture executives such as Jen Kha, a managing partner at Andreessen Horowitz, argue that the mega-IPOs have changed the prospects for private investments more broadly and that VC should play a bigger role, particularly when compared with buyout funds. These have “historically captured a much larger share” of investment, Kha wrote recently, “but venture outcomes are simply changing the math.”

It doesn’t hurt the VC case that some buyout firms have suffered considerable pain from a favourite investment strategy, rolling up software companies with cash flows that once seemed impregnable but no longer. In the words of Kha: With AI threatening considerable disruption in the corporate world, “there may be a whole lot of pain in PE”.

However, the hope that investors may have been rethinking their commitments to venture is belied by the figures. Even as investment in AI boomed, the new cash being put into venture funds tailed off after 2021. With so much money already locked up in older companies, many investors have not had the cash to recycle into new investments and have not been prepared to increase their overall allocations.

The exception has been for a small number of very large venture firms that have taken to raising multibillion-dollar amounts, often for a mix of early- and growth-stage funds. Just three firms — Andreessen, Founders Fund and Thrive Capital — raised about $25bn between them in the first six months of this year, almost a third of all the new money that found its way into venture in the US.

Recommended

When the profits from the AI mega-IPOs are counted, the narrowness of the winners’ circle is likely to increase the pressure to get access to funds such as these.

Whether this permanently changes the dynamics of VC is another matter. There are good reasons to think that the gains from the AI boom will broaden out from here. A handful of chipmakers, model-builders and cloud platform companies have benefited disproportionately from the early build-out, but there will be plenty of room for the many other start-ups building out the broader technology ecosystem around AI and the new applications that turn it into something useful for customers.