Databricks confidentially files for IPO with SEC, eyes 2026 listing amid AI spending surge
Enzo Reno ·
Databricks has confidentially submitted a draft registration statement for an initial public offering to the U.S. Securities and Exchange Commission on Wednesday, according to people familiar with the matter. The San Francisco company is targeting a listing later in 2026, subject to market conditions, as enterprise demand for AI infrastructure and data tools remains elevated.
The move comes as the U.S. IPO window continues to reopen following 2024 tech debuts from Reddit, Astera Labs, and Rubrik, which broadened investor appetite for profitable growth and AI exposure. A successful debut would be a barometer for late stage enterprise software valuations tied to generative AI workloads.
Why it matters
Databricks competes directly with Snowflake for data platform spend and has pushed into generative AI with its DBRX model, released in March 2024, and integrations across Microsoft Azure, Amazon Web Services, and Google Cloud. The company last raised 500 million dollars in September 2023 at a 43 billion dollar valuation, with NVIDIA joining the round, and said then that it had surpassed a 1.5 billion dollar revenue run rate and served more than 10,000 customers.
A listing would give Databricks a public equity currency to fund acquisitions after deals for MosaicML in 2023 for 1.3 billion dollars and data replication startup Arcion in 2023 for about 100 million dollars. It would also intensify competitive pressure on Snowflake, which named Sridhar Ramaswamy as chief executive in 2024 and has been emphasizing AI driven search and application workloads.
What the filing signals
A confidential submission allows the SEC to review financials and risk disclosures before a public registration is posted. Large issuers typically go through multiple amendment cycles over several months, with banks finalizing valuation ranges closer to the roadshow, according to capital markets lawyers and bankers who advise on technology listings.
Databricks has highlighted rising AI infrastructure costs, including GPU availability and power constraints, as a customer planning issue at industry events, while partners have leaned on its lakehouse architecture to consolidate data and model operations. Public market feedback will likely focus on gross margin sustainability as AI workloads scale, the mix of consumption based revenue, and competition from cloud provider native services.
If the deal proceeds on the current timetable, it would be a defining test of how public investors price AI platforms that straddle data management and model tooling, and it could set reference points for other large private companies in AI and fintech that have been weighing listings, including Stripe and Cohere.