Despite software-as-a-service making a (modest) comeback, investors are still struggling to determine the value of these pre-AI software companies that dominated venture portfolios before the AI boom. 

The problem isn’t just that AI companies are attracting all the capital, but that it’s hard to assess which SaaS companies will retain an advantage in an AI-driven market and which will become obsolete.

“I genuinely worry about the pre-AI venture portfolio,” said one healthcare CIO who oversees roughly $11 billion in assets. “Prior to 2022, there was a lot of SaaS; now, everyone’s like, ‘I don’t know how much this is worth.'” 

SaaS companies saw a boom after the pandemic boosted the need for online services, bringing in a record $297 billion of global investor capital in 2021, according to SaasRise. But after ChatGPT’s launch in late 2022, VC funding pivoted almost entirely to AI, bringing that number to $200 billion in 2025, down nearly 33 percent from its 2021 peak. Meanwhile, investors have pumped $222 billion into U.S. AI companies—a 204 percent increase since 2022. 

Now, AI companies account for nearly a third of all global venture deals completed in 2025. And they continue to dominate the venture landscape in 2026: Data from Crunchbase show that OpenAI and Anthropic accounted for 43 percent of all startup funding in the first half of this year (both companies have much-anticipated IPOs on the way).

Although software VC deals are rebounding, they’re nowhere near their pre-AI peak. Deal value rose nearly 13 percent year-over-year, and volume rose 12 percent to an estimated 9,200 deals — the first year-over-year increase since 2021. Still, that's below the 16,446 deals reached at the sector’s 2021 peak.

“If you look at 2020, 2021 vintages, the returns are really concerning,” said Joshua White, director of private equity at OCIO provider Hirtle & Co. (formerly Hirtle Callaghan). “The median return is significantly below the historical average.” 

White explained that many software companies once commanded high valuations because they were growing 80 percent annually. Now that's declined to 15 percent. While that's still healthy growth for most businesses, it's not enough for growth investors.

After AI threw off valuations for the past few years, enterprise software was hit the hardest: CNBC reported that 75 SaaS firms appeared on PitchBook's list of fallen unicorns (startups once valued at $1 billion or more). That's twice as many as fintech companies, the next-largest category.

What’s happening in AI and software has occurred before in previous waves, including the rise of mobile and the Internet—but not necessarily to this degree. Software stocks saw a huge selloff in February, driven by fears of AI disruption and months of underperformance, with the S&P 500 software and services index losing nearly $1 trillion in value.

“It’s not unique, just broad in terms of number of use cases that Anthropic or OpenAI can displace,” White said. “The workflow SaaS companies that are really threatened are the ones where there’s a human in the beginning and a human in the end with the workflow in the middle.” (These include translation or transcription services.)

For many pre-AI software companies, the impact of AI depends on their business model. Some, like mobile banking app Chime, have seen little effect, while others, including data and analytics firm Databricks and trucker safety equipment manufacturer Motive, have benefited from bolting on AI functionality to their services.

Scott Martin, head of global co-investments at Cambridge Associates, shared this sentiment. The investment consultant noted that the leading AI companies will disrupt multiple industries for a while but eventually will benefit these software firms. 

While the future of many SaaS startups may be in limbo, Martin thinks the crisis is overstated. “Despite what everyone says, AI is software, and if you’re not using AI as a software company, you will fail,” Martin explained.