Data and artificial intelligence (AI) company Databricks has entered into a definitive agreement with MosaicML for US$1.3 billion ($1.76 billion) on June 26. The deal was paid with a mix of cash and stock in Databricks.
The news comes after the company crossed its US$1 billion revenue goal for the FY2023 ended Jan 31, on June 14.
MosaicML is a generative AI platform that enables enterprises to build their own AI. It is known for its state-of-the-art MPT (or MosaicML pre-trained transformer) large language models (LLMs).
The acquisition comes at a time when AI is changing how people work and businesses operate.
“Every organisation should be able to benefit from the AI revolution with more control over how their data is used. Databricks and MosaicML have an incredible opportunity to democratize AI and make the Lakehouse the best place to build generative AI and LLMs,” says Ali Ghodsi, co-founder and CEO of Databricks.
The acquisition also “just makes a lot of sense,” he adds at a press conference in San Francisco, noting that in the last six or seven months, every large enterprise that he’s talked to “always ends up going towards generative AI”.
See also: Microsoft warns other firms of Russian-sponsored group in email hacking
“Their ability to train models, our ability to have thousands of customers that have already trusted us with their data, combined with our go-to model, we think it’s going to be a force multiplier, that’s why we’re excited,” he says.
According to Databricks, the company’s lakehouse platform, combined with MosaicML’s technology, helps lower costs for its customers while offering them a simple and fast way to retain control, security and ownership over their data. The move will cost companies that train and use LLMs thousands of dollars instead of millions.
“Databricks’ unified data and AI platform combined with MosaicML's generative AI training capabilities will provide a platform robust enough to serve the world's largest organizations and flexible enough to address a broad range of AI use cases,” reads the statement released by Databricks on June 26.
See also: Microsoft, Amazon and Google are kingmakers for AI start-ups
On the amount, CEO Ghodsi explains that the sum is justified with a “conservative” valuation of 10 to 20 times of annual recurring revenue (ARR) and is likely to be made back in less than a year.
He explains: “Let's say a company only gets 10x multiple on the revenue, which is conservative, because [there are] other data companies right now in the public markets that are getting close to 20x. To justify US$1.3 billion at a 10x multiple, you would have to get US$130 million.
“[MosaicML] were at US$1 million at the beginning of the year [and] now they are at US$20 million. You can forecast, how long it will take for them, independently at their growth rate to get to US$130 million. And combined with Databricks, with our full platform and our servers, how long will it take us to get to US$130 million? I’d say less than a year.”
He adds: “[But] this is not why we’re buying the company. That's absolutely not the case. We are super excited about democratizing AI, to all the startups. We think this is going to be massive over the next five years.”
Following the acquisition, the entire MosaicML team – including its research team – will join Databricks.
“At MosaicML, we believe in a world where everyone is empowered to build and train their own models, imbued with their own opinions and viewpoints — and joining forces with Databricks will help us make that belief a reality,” says Naveen Rao, Co-Founder and CEO, MosaicML.
“We started MosaicML to solve the hard engineering and research problems necessary to make large-scale training more accessible to everyone. With the recent generative AI wave, this mission has taken centre stage. Together with Databricks, we will tip the scales in the favour of many — and we’ll do it as kindred spirits: researchers turned entrepreneurs sharing a similar mission,” he adds.
