Without a doubt, artificial intelligence will change our world a lot in the long run. But for now, we may be living in an AI bubble.
Those looking for evidence of this might cite the news of Cognition Labs seeking a $2 billion valuation, as reported by Wall Street Journal on Saturday.
Founded in November, Cognition Labs produces Devin, which it describes as the “first fully autonomous AI software engineer.” He generated no real revenue. He launched Devin this month.
Earlier this year, the startup raised $21 million in a deal worth $350 million. He then rejected offers worth $1 billion. Now, according to the magazineis in talks with investors about a deal that would value it at up to $2 billion.
That’s a staggering figure for a new business. But that’s not all that shocking in today’s AI space. Perplexity, an AI search startup challenging Google, secured funding a few weeks ago valuing it at $1 billion, up from $520 million a few months earlier, with Amazon founder Jeff Bezos among the supporters. Mistral, a French startup founded just over a year ago, reached a $2 billion valuation in December.
“Every bubble has a compelling story”
Each of these startups could justify their high valuations. But as more and more AI ventures fetch improbably large sums from investors who spread their bets, the sense of a bubble is growing among some observers.
Albert Edwards, chief global strategist at Société Générale, is among the skeptics.
“Every bubble has a compelling story,” he wrote in a note this week. “The current narrative is centered on anticipating an AI-driven surge in corporate profits to fully justify current stratospheric valuations. Those of us who lived through the late 90s [tech] bubble we’ve heard it all before and we roll our eyes.
As for Devin, “many companies are working on some variation of this idea,” noted venture capitalist David Sacks on a recent episode of Full podcast. While he likes the firm’s “agent-first” approach to generating new software projects, “where I think this becomes much more complicated and much more difficult is when you work on existing code bases,” a challenge that others AI startups are tackling.
One advantage of Devin, he added, is that “he’ll do a really good demo.”
Of course, it remains to be seen whether the compelling demonstrations that amaze investors today will translate into thriving companies in the years to come. In any case, today’s eye-popping valuations for unproven startups will likely be remembered.