Venture capital is always chasing some hot theme. Five years ago, it was software-as-a-service. A couple years later, it was Web3. Now, it’s artificial intelligence.
But while VC investors chase trends, they still typically diversify. Currently, they are monomaniacally chasing AI above all others in a way the investment and tech industries have not seen before.
VCs are putting a big percentage of new money into AI, increasing concentration risk for investors. Stocks are also dominated by AI, with roughly a third of the S&P 500’s value coming from large AI-focused technology stocks.
“I don’t think we’ve ever seen the kinds of concentration in themes that we’ve been seeing in venture portfolios,” Brad Conger, chief investment officer for Hirtle Callaghan told Institutional Investor. “It’s like nothing else. It’s all in AI.”
According to Conger, VC firms are currently not backing retail startups, consumer startups, or application software. Instead, he sees almost “every dollar going into venture is going into AI.” He added that even if the companies are not explicitly labeled AI, their business models anchor on some AI angle.
AI is grabbing the lion’s share of capital at a time when VC funding is on track to hit a multi-year low, further amplifying the concentration. Data from PitchBook show that so far this year, only 823 venture funds have raised over $80 billion — a severe drop from 2022, when 4,430 funds raised roughly $412 billion.
As of September 30, AI startups have attracted nearly $193 billion — or 53 percent — of the nearly $367 billion in venture capital deals this year (AI companies accounted for 65 percent of VC capital and 35 percent of the number of deals during the first half of 2025).
Conger noted that “there may be under-reporting of AI-enabled vs AI business” due to GPs “not wanting to appear to be casting all their bets on the same theme.”
One allocator at a health care system said that with so many venture portfolios directly leveraged to AI in some way, it could even affect other non-AI companies in the private equity portfolio (they suggested that many of these older companies could be figuring out how to become more productive and competitive through AI).
This intense focus is creating two distinct venture markets: one for AI and one for everything else. What this concentration of capital means is that even as the overall VC market shows signs of life, the recovery is lopsided, driven overwhelmingly by a single sector.
The topic of discussion for any investment committee will be what to do with their venture portfolios, which means AI. So, Conger believes how allocators invest in AI “will be the big issue in terms of planning for 2026.”
“VCs Are Doing What They Should Do”
While allocators with whom II spoke agreed that most of the venture money is going into AI, they aren’t saying that’s necessarily a bad thing. The healthcare investment chief noted that “some people will be huge benefactors, and some people might be in the path of creative destruction,” like what happened with the Internet and other technology waves.
“We think about valuations, but we're not market timers, so we don't try to time based on that,” the allocator said. “But our notion is to lean into quality as much as we can when we see valuations creeping up, like they have been right now.”
Paul O'Brien, trustee and investment committee member at the $12 billion Wyoming Retirement System, said he hasn’t seen “the concentration of economic risk and opportunity in a single investment theme. At least since railroads, and VC wasn’t around then.”
As O’Brien explained it, a VC portfolio is a package of call options, which are worth more when volatility rises. Since volatility is far higher in AI than in other sectors, he expects “some huge winners in current VC vintages.”
“So, VCs are doing what they should do,” O’Brien added.
With that, O’Brien explained that allocators should understand that VC is giving a concentrated bet on AI, so they need to seek diversification elsewhere. In addition to diversifying, CIOs should also develop an overall portfolio construction strategy around AI (“Do you want your exposure through VC or through Nvidia and Alphabet?” he posed).
“If you have access to some experienced VC investors I’d say go for it. But don’t allocate capital you can’t afford to lose,” O’Brien said before adding: “Get some hedges. There will be a bust at some point, and you’ll need to cover your liabilities.”
