Databricks Hits 188 Billion Dollar Valuation and Nobody Should Be Surprised

There is a particular kind of company that thrives not by being the loudest name in a space, but by being genuinely indispensable to the people building it. Databricks is that company. This week, the data and AI platform closed a new funding round that values it at 188 billion dollars, cementing its place as one of the most valuable private companies on the planet.

To put that number in context: 188 billion dollars is more than Ford, Volkswagen, and General Motors combined. For a company most people outside the enterprise software world have never heard of, that is a staggering position to be in.

What Databricks Actually Does

Founded in 2013 by the creators of Apache Spark at UC Berkeley, Databricks built its name on making it easier for large organisations to process and analyse enormous datasets. Over time, it evolved into a full data intelligence platform, helping companies store, manage, and now train AI models on their proprietary data.

That last part is the key to understanding why investors keep piling in. Every major enterprise that wants to build its own AI capability, rather than simply plugging into OpenAI or Google, needs infrastructure for handling data at scale. Databricks sits right at that bottleneck. It is the picks-and-shovels play of the AI gold rush, and it has been executing quietly and consistently while flashier AI startups grabbed headlines.

The company crossed 3 billion dollars in annualised revenue earlier this year, making it one of the fastest-growing enterprise software companies in history. It is also, notably, profitable on an adjusted basis, which puts it in a very different category from many of the AI darlings burning cash to chase growth.

The Second Act Nobody Saw Coming

What makes Databricks interesting right now is that it is not resting on data infrastructure. The company has been making aggressive moves into AI model development, most notably through its acquisition of MosaicML in 2023 for 1.3 billion dollars. That deal brought serious model training capability in-house and gave Databricks the ability to offer customers end-to-end AI development, from raw data to deployed model.

Its open-source DBRX model, released in early 2024, was a statement of intent. It was not trying to compete with GPT-4 on benchmark bragging rights. It was designed to be fine-tuned on a company's own data, which is exactly what enterprise customers actually want.

More recently, Databricks has been building out agentic AI tooling, helping organisations move from static models to systems that can take actions, query data, and reason across multiple steps. This positions it well for what most analysts expect to be the next major enterprise AI spending wave.

Why the Valuation Makes Sense

There is a reasonable argument that 188 billion dollars is a lot of money for a company that most consumers cannot name. But enterprise software has always been valued on different terms, and Databricks earns every penny of its premium on a few key metrics.

First, switching costs are enormous. Once a company has built its data pipelines, AI workflows, and model training infrastructure on Databricks, migrating away is a multi-year project. That kind of lock-in is worth a great deal to investors.

Second, the addressable market is genuinely vast. Every large organisation on earth is trying to figure out how to use AI on its own data without sending that data to a third party. Databricks is one of a very small number of credible answers to that problem.

Third, and perhaps most importantly, Databricks has a track record of shipping things that work. That sounds like a low bar, but in enterprise AI right now, it is actually quite rare.

An IPO has been long anticipated. The company filed confidentially for a public listing earlier this year, and a valuation this size suggests it is getting serious about that path. When it does go public, it will be one of the most closely watched listings in years.

For now, Databricks is proof that you do not need to be building the most exciting AI model in the world to build extraordinary value. Sometimes being the most necessary thing in the room is more than enough.