Navigating the Risks: Why Investing in Growth-Stage AI Startups Is Becoming More Complex

The allure of investing in AI startups is undeniable, but the landscape is shifting, presenting both exciting opportunities and increased risks. Established giants like OpenAI, Microsoft, and Google are rapidly expanding their capabilities, potentially overshadowing smaller companies. Simultaneously, new AI startups are achieving growth-stage status at an unprecedented pace.

Defining the “growth stage” for AI startups is becoming increasingly nuanced. According to Jill Chase, partner at CapitalG, companies as young as one year old are reaching significant milestones, boasting tens of millions in annual recurring revenue (ARR) and valuations exceeding $1 billion. Chase shared her insights at TechCrunch AI Sessions.

“On one hand, that’s really exciting. It represents this brand new trend of extremely fast growth, which is awesome,” Chase noted. “On the other hand, it’s a little bit scary because I’m gonna pay at an $X billion valuation for this company that didn’t exist 12 months ago, and things are changing so quickly.”

The rapid pace of innovation creates uncertainty. “Who knows who is in a garage somewhere, maybe in this audience somewhere, starting a company that in 12 months will be a lot better than this one I’m investing in that’s at $50 million ARR today,” Chase added. “So it’s made growth investing a little confusing.”

To navigate this complex environment, Chase emphasizes the importance of understanding the market category and assessing the founder’s ability to adapt quickly and anticipate future trends.

She highlighted AI coding startup Cursor as an example of a company that capitalized on a timely use case: AI code generation. However, maintaining a competitive edge requires continuous innovation. “There will be, by the end of this year, AI software engineers,” Chase predicted. “In that scenario, what Cursor has today is going to be a little less relevant. It is incumbent on the Cursor team to see that future and to think, okay, how do I start building my product so that when those models come out and are much more powerful, the product surface represents those and I can very quickly plug those in and switch into that state of code generation?”