Why investing in growth-stage AI startups is getting riskier and more complicated

The betting on AI startups is so exciting – or even more risky has never been. Responsibles like OpenAI, Microsoft and Google are fast scales to consume many offers of small companies. At the same time, the new startups histor has reached the level of growth much faster than them.

However, defining the “stage of growth” in AI startups is not so cut off today.

Capitalog’s partner Jill Chase says on the stage TechCrunch AI Sessions That he is watching more companies that are only one year old, yet the annual repetition has reached a few million in the annual repetition and more than $ 1 billion evaluation. Although these companies can be defined as mature due to their evaluation and earning production, they often lack the necessary protection, recruitment and executive infrastructure.

“On the one hand, it is really exciting IT it presents the new trend of this brand of very fast development, which is great,” “on the other hand, it is a bit scary because I will pay a $ x billion evaluation for this company 12 months ago and things are changing so fast.”

Chase also said, “Who knows who in the garage knows, perhaps somewhere in these audiences, starting a company that I am investing in $ 50 million today in 12 months will be much better than I am investing today,” Chase said. “So it has made a little confusing invested in the growth.”

To cut the noise, Chase said that it is important for investors to feel good about the department and “the founder’s ability to adapt very quickly and to the corner” “” “

He mentions that the AI ​​Coding Startup Cursor is a great example of a company that jumps in the right use of the AI ​​code generation jumps jump jumps jump jumps jump jump jumps jump.

However, the cursor has to work to maintain its edge.

“By the end of this year, AI software engineers will be,” Chess said. “In that scene, the cursor is going to be a little less relevant today. The cursor team has to see and think of that future, okay, how do I start making my product so that when these models come out and stronger, I represent them and can quickly plug them and switches them in the codes generation.”