As SpaceX prices its blockbuster listing this week and OpenAI and Anthropic line up behind it, a fourth name keeps appearing on every AI IPO watchlist — and it is the odd one out for a simple reason. Databricks, the data and AI software company, is the only profitable company in the entire AI IPO pipeline. In a cohort defined by enormous valuations and enormous losses, Databricks generates positive free cash flow, and that single fact reframes how to read the whole wave.
For an investor sizing up the most crowded IPO calendar in years, Databricks is the control group: the one AI-era candidate whose economics already work, against which the cash-burning giants can be measured.
What Makes Databricks Different?
Databricks has the growth profile investors expect from an AI darling, plus the financials they usually don't get. The company crossed a $5.4 billion revenue run-rate growing more than 65% year over year in the fourth quarter of 2025, and it has been free-cash-flow positive over the trailing twelve months. Its net revenue retention rate sits above 140%, meaning the average existing customer spends more than 40% more each year than the year before — the hallmark of software that becomes more entrenched the longer a company uses it.
That combination is rare. For comparison, the publicly traded data company Snowflake reported a roughly $1.3 billion net loss in its most recent fiscal year on about $4.7 billion in revenue. Even an established, listed peer in the same business loses money, which underscores how unusual it is for a company growing this fast to also be cash-generative.
How Does It Compare to the Rest of the Wave?
The significance of Databricks' profitability comes into focus next to the companies it is grouped with. SpaceX is going public this week at a valuation near $1.77 trillion — roughly 94 times sales — and OpenAI and Anthropic, valued at about $852 billion and $965 billion respectively, are both deeply unprofitable, pouring capital into compute and research far faster than revenue comes in. The marquee AI IPO class is, in effect, asking public investors to fund years of losses on the promise of eventual dominance.
Databricks is asking something different. It is the rare AI-era company that has already shown it can grow quickly and make money at the same time, which makes it a kind of test case for the whole debate about whether AI valuations are rational. If the bubble skeptics are right that most of these companies will never earn back what is being spent, Databricks is the exception — the one whose unit economics already work.
Why Does Databricks Make Money When the Others Don't?
The answer comes down to where each company sits in the AI stack. OpenAI and Anthropic build frontier models, which requires spending billions on training compute before revenue catches up; SpaceX is pouring capital into rockets, satellites, and now orbital data centers. Databricks, by contrast, sells the data platform that enterprises use to organize their information and build their own AI on top of it — a software business with the high gross margins and usage-based pricing software is known for.
In practical terms, Databricks pioneered the "lakehouse," an architecture that merges the flexibility of a data lake with the structure of a data warehouse, giving companies one place to store, govern, and analyze the data AI models need. It monetizes the AI boom without bearing the heaviest cost of it: rather than training the models, it sells the rails for the data layer and charges by usage. That is why its growth shows up as profit rather than as a widening loss — and why a slowdown at the model labs would hurt it far less than it would hurt them.
Is Profitable the Same as Cheap?
No — and that is the qualifier for anyone tempted to read "profitable" as "safe bet." Databricks is still valued like a hyper-growth company. It was worth $134 billion in a December 2025 funding round, and according to reporting from The Information it is in talks to raise again at a valuation between $165 billion and $175 billion — roughly 30 times its revenue run-rate. Profitability removes one of the biggest risks hanging over its peers, but it does not make the stock inexpensive, and a rich multiple still demands years of strong execution to justify.
Notably, Databricks is also in no hurry. As of mid-2026 it had not filed an S-1, the registration that precedes a US listing, and analysts broadly expect a filing around the third quarter of 2026 with a debut in late 2026 or early 2027 — behind SpaceX, OpenAI, and Anthropic. That patience is itself a sign of strength: a cash-generative company can choose to list from a position of strength rather than out of a need to raise money, and Databricks has even taken on debt to fund itself rather than rush to market.
What Does It Mean for the AI IPO Story?
For investors trying to make sense of the calendar, Databricks is a useful reference point. The headline names are bets on the future, asking the market to finance enormous spending today against the promise of transformative returns later; Databricks is a bet on a present-tense business that already works, in the unglamorous but essential layer of data infrastructure.
How the market treats it will be revealing. If public investors reward Databricks' profitability with a premium, it would suggest the appetite for AI exposure is becoming more discerning, favoring proven economics over pure growth. If the bigger, cash-burning names command higher multiples anyway, it would confirm the market is still paying most for the boldest promises rather than the soundest numbers. Either way, in a wave of AI listings built on the willingness to lose money now, Databricks stands out as the one that doesn't have to. This article is not investment advice.
Frequently Asked Questions
Is Databricks profitable?
Databricks has been free-cash-flow positive over the trailing twelve months, making it the only profitable company in the AI IPO pipeline. It crossed a $5.4 billion revenue run-rate growing more than 65% year over year in late 2025, with a net revenue retention rate above 140%.
When will Databricks IPO?
As of mid-2026, Databricks had not yet filed an S-1, the registration document that precedes a US listing. Analysts broadly expect a filing around the third quarter of 2026 and a public debut in late 2026 or early 2027 — later than SpaceX, OpenAI, and Anthropic.
How is Databricks valued?
Databricks was valued at $134 billion in a December 2025 funding round and, according to The Information, is in talks to raise again at $165 billion to $175 billion — roughly 30 times its revenue run-rate. So while it is profitable, its valuation is still that of a hyper-growth company.
Why is Databricks profitable when OpenAI and Anthropic are not?
Databricks sells a high-margin data-platform software business and charges customers by usage, monetizing AI adoption without training frontier models. OpenAI and Anthropic spend billions on compute to build models before revenue catches up, which is why they post large losses while Databricks generates cash. This article is not investment advice.
ⓒ 2026 TECHTIMES.com All rights reserved. Do not reproduce without permission.
