Editorial illustration for Databricks CEO to Snowflake: 'You’re Not Doing This AI Stuff
Databricks CEO Challenges Snowflake on AI Strategy
Databricks CEO to Snowflake: 'You’re Not Doing This AI Stuff
Databricks CEO Ali Ghodsi picked a fight with Snowflake on X this week, and he did it in public. The exchange started after Snowflake CEO Sridhar Ramaswamy pushed Databricks to open its books, pointing to a Ben Horowitz post titled "Databricks & the Future of the Firm" and writing simply, "Release all the numbers." Ramaswamy wanted specifics: whether Databricks' reported 80% year-on-year revenue growth would shrink to 65% once pass-through inference revenue got stripped out, how much of that growth came from AI-native customers versus legacy software clients shifting to service-based billing, and what the company's free cash flow and remaining performance obligations actually looked like.
Rather than answer with figures, Ghodsi turned the conversation toward AI infrastructure. He insisted the two companies aren't rivals, calling them "complementary," before pivoting into a sales pitch aimed squarely at his competitor's own platform. The back-and-forth, playing out in full view of both companies' customers and investors, turned a pointed question about revenue transparency into something closer to a product demo. What Ghodsi said next is where things got pointed.
Databricks CEO Ali Ghodsi took a jab at rival Snowflake over its AI strategy, telling the company to consider buying AI tokens through Databricks after Snowflake CEO Sridhar Ramaswamy questioned the data software company’s revenue growth.
Why this matters
Public spats between CEOs make good screenshots, but this one exposes something founders and data teams should actually track: neither Databricks nor Snowflake has published hard numbers on what portion of their revenue comes from AI workloads versus legacy data warehousing and lakehouse queries. Ghodsi's jab and Ramaswamy's pointed "I was asking about the revenue, not contributing to it" both dodge the real question Horowitz's piece raised, which is whether these platforms are genuinely embedded in enterprise AI stacks or just riding the hype cycle in their marketing. For anyone choosing infrastructure right now, that distinction matters more than Twitter banter.
If you're building on either platform, push your sales rep for actual usage breakdowns before you commit spend. The fact that two CEOs are trading jabs instead of disclosures suggests the numbers, when they do surface, might complicate the narrative each company is selling. Watch for Snowflake's and Databricks' next earnings calls.
That's where this argument actually gets settled, not on X.
Related Reading
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups DeepSeek Seeks More Capital Weeks After USD 7B Funding Round
From Archives Sarvam, Krutrim Secure 68% of India's USD 687 Million AI Funding
From Archives HCLTech Launches Nano GCC AI Model With AWS From Lucknow Hub
Common Questions Answered
What did Databricks CEO Ali Ghodsi claim about Snowflake's AI strategy?
Ali Ghodsi publicly criticized Snowflake on X, telling the company that it's not adequately pursuing AI initiatives and suggesting that Snowflake should consider buying AI tokens through Databricks instead. This criticism came in response to Snowflake CEO Sridhar Ramaswamy questioning Databricks' reported 80% year-on-year revenue growth figures.
Why did Snowflake CEO Sridhar Ramaswamy challenge Databricks to release financial numbers?
Ramaswamy questioned whether Databricks' claimed 80% year-on-year revenue growth would actually shrink to 65% once pass-through inference revenue was excluded from the calculations. He wanted transparency on how much of Databricks' growth was genuinely attributable to core business performance versus inflated by AI token pass-through revenue.
What is the underlying issue that the Databricks-Snowflake dispute exposes?
Neither Databricks nor Snowflake has publicly disclosed what portion of their revenue comes from AI workloads compared to legacy data warehousing and lakehouse queries. This lack of transparency makes it difficult for founders and data teams to understand how much of these companies' growth is actually driven by AI capabilities versus traditional data platform services.
How did Ramaswamy respond to Ghodsi's suggestion about buying AI tokens?
Ramaswamy deflected the suggestion by responding that he was asking about revenue transparency, not offering to contribute to Databricks' revenue. His response suggested that Ghodsi avoided addressing the core question about revenue breakdown that was raised in the Ben Horowitz analysis.
Further Reading
- Databricks CEO Ali Ghodsi on Snowflake rivalry and the 'why' behind Databricks’ latest billion-dollar deal - Fortune
- Databricks CEO Says Rivalry Spurred Executive Swap at Snowflake - Bloomberg Law
- Snowflake faces intensifying competition from Databricks, but Jefferies sees room for both - Yahoo Finance
- On theCUBE: Snowflake, Databricks and more battle for control of AI stack - SiliconANGLE
