Databricks has just closed one the largest funding rounds in history, raising an astounding $10 billion of new capital. Investors in technology were quick to ask how this would affect the highly anticipated IPO. Databricks CEO Ali Ghodsi revealed his reasoning for waiting until 2025 at an event in San Francisco.
This year was an electoral year. We wanted some stability – people are worried about inflation, interest rates. We decided to wait, because it would be stupid to IPO in this year. Axios AI Summit. “The earliest possible IPO would be in next year. And then there are lock-up periods. So it would be just too long for employees to gain liquidity.”
Databricks This “Series J”, is used to allow early employees to cash out and continue growth. While 2024 was uncertain, the IPOs by ServiceTitan and Reddit have been largely successful.
Why risk it when Databricks can raise so much money?
Ghodsi stated that this latest round may have been almost double the amount of money raised just before. We knew investors wanted to get in but the craze led Databricks’ share price to rise. Ghodsi says the data analytics company initially tried to raise $3 billion to $400 billion in this round. Press reports about their fundraising sparked a flurry of interest.
I saw this Excel sheet, where they keep track of all the investors. Ghodsi said, “It was $19 billion in interest and I almost fell out of my chair.” “And we still hadn’t talked to everyone.” I was like ‘Oh my god, that’s an enormous amount of numbers. So we actually moved up the price.’
Despite the impressive fundraise Ghodsi doesn’t rule out a Databricks initial public offering in 2025. He said that it could also be in 2026. He said that although it is less important than it was in the past 10 to 15 years to go public, the company still wants to do it. Ghodsi, however, is not trying to rush an IPO in order to avoid the “AI bubble,” which he called, from bursting.
I mean, it’s the peak of the AI bubble. Ghodsi said that it doesn’t take an expert to realize that a startup with only five people, and no product, innovation, or IP — just a few recent graduates — is [is not] valued at hundreds of millions, if not billions. “These startups are valued at billions of dollars, and they have nothing. That’s a bubble.” The Databricks CEO did not specify which startups he was referring to. However, we’ve seen a number of AI unicorns in 2018. Ghodsi isn’t worried about any of this, though. He believes that his company and its valuation will stand the test. He believes his company has already won its first major battle against another data analytics startup called Snowflake.
Ghodsi confirmed that they had a program named SnowMelt. Reports of an initiative by Databricks to steal Snowflake’s businesshave been made. “We demonized Snowflake, but that is behind us.”
This effort to demonize Snowflake cost Databricks $2 billion for a tiny startup named Tabular. Snowflake reportedly also wanted to buy Tabular even though it only had $1 million in annual revenue recurring at the time.
Now Databricks has products that compete with enterprise giants such as Salesforce and Microsoft. Ghodsi believes that data and AI will continue play a more important role in the lives of people every year. He thinks that his company is well positioned to fill this niche. Maxwell Zeff, a senior reporter for TechCrunch who specializes in AI and emerging technology, is
Maxwell Zeff. Zeff covered the rise and fall of AI, as well as the Silicon Valley Bank Crisis, for Gizmodo and MSNBC. He is based out of San Francisco. When he is not reporting, you can find him hiking, biking and exploring the Bay Area food scene.
View Bio