Databricks Bets $100M on OpenAI to Boost Enterprise AI
▼ Summary
– Databricks is integrating OpenAI’s models including GPT-5 into its platform and Agent Bricks product through a $100 million multi-year deal.
– This partnership highlights the competitive race to incorporate generative AI tools that can securely use corporate data into enterprise software stacks.
– Agent Bricks now offers OpenAI’s latest models, with GPT-5 as a flagship option, allowing customers to build AI apps on their enterprise data via SQL or API.
– The deal involves a guaranteed $100 million minimum payment from Databricks to OpenAI, regardless of whether the models generate that much revenue.
– Databricks has already seen significant customer demand for this integration, including from companies like Mastercard.
In a major strategic move, Databricks is committing $100 million to integrate OpenAI’s advanced artificial intelligence models directly into its data platform. This multi-year partnership aims to significantly enhance the capabilities of Databricks’ AI product, Agent Bricks, by incorporating models like the highly anticipated GPT-5. The central goal is to provide enterprise clients with powerful generative AI tools that can securely interact with their proprietary corporate data, addressing a rapidly growing market demand.
This development underscores the intensifying competition among tech firms to embed sophisticated AI into the enterprise software stack. Agent Bricks empowers organizations to develop custom AI applications and agents leveraging their own data, and it now offers OpenAI’s latest models as part of its toolkit. Customers can access these models through SQL queries or API calls, with GPT-5 positioned as a flagship option. This integration builds on a previous update from nearly two months ago, when Databricks added OpenAI’s open-weight models, gpt-oss 20B and gpt-oss 120B, to its platform.
A key feature of Agent Bricks is its ability to evaluate the performance of different AI models on specific tasks. The system can then fine-tune the models accordingly, enabling businesses to achieve more precise and tailored results for their unique needs. Brad Lightcap, OpenAI’s Chief Operating Officer, emphasized the strategic value of the partnership, stating that it places their most advanced AI where secure enterprise data already resides, simplifying the process for companies to experiment with, deploy, and scale impactful AI agents.
The financial structure of the deal involves a significant commitment from Databricks. The company has agreed to pay OpenAI a minimum of $100 million over the agreement’s term, regardless of whether the usage of OpenAI’s models actually generates that level of revenue. This arrangement provides OpenAI with a predictable income stream, which is particularly valuable as it invests heavily in expanding its data center infrastructure. For Databricks, however, it represents a calculated risk, betting that enterprise customer adoption will be strong enough to justify the guaranteed payment. The specific duration of the agreement was not publicly disclosed.
This partnership mirrors a similar $100 million, five-year revenue target agreement that Databricks established with Anthropic earlier this year. According to a Databricks spokesperson, the company is already witnessing substantial customer interest in having native access to OpenAI’s models on its platform, with major clients like Mastercard expressing significant demand.
(Source: TechCrunch)
