Artificial Intelligence

Databricks CEO: AI is Making Traditional SaaS Interfaces Irrelevant

by Sakshi Dhingra - 4 days ago - 2 min read

In a bold assessment of the enterprise software landscape, Databricks CEO Ali Ghodsi has declared that while Software-as-a-Service (SaaS) isn’t dead, it is rapidly approaching a state of irrelevance. Speaking on the heels of the company’s massive $134 billion valuation, Ghodsi argues that the traditional “point-and-click” era of software is being swallowed by the "prompt-and-execute" power of Artificial Intelligence.

According to Ghodsi, the threat to incumbents like Salesforce or SAP isn't that companies will suddenly delete their databases. These "systems of record" are too deeply embedded in corporate infrastructure to be easily replaced. Instead, the danger lies in the "interface layer."

“Everybody is like, ‘What’s AI going to do with all these companies?’” Ghodsi noted. For Databricks, the answer has been an explosion in usage. The company recently reported a $5.4 billion revenue run rate, fueled largely by AI-native tools like Genie—an interface that allows non-technical users to query massive datasets using simple natural language.

The "moat" that once protected SaaS giants—years of user training and specialized certifications—is evaporating. When an AI agent can file a ticket, pull a financial report, or adjust a sales quota via a chat box, the specific menus and dashboards of the underlying software become invisible.

“Why would you move your system of record? It’s hard to move,” Ghodsi explained. But as AI agents become the primary way employees interact with data, the "SaaS" part of the business becomes mere plumbing. The value, and the revenue, are shifting to the platforms that power the intelligence.

With a fresh $5 billion in capital and AI-driven products already contributing $1.4 billion to its bottom line, Databricks is positioning itself as the foundation for this new era. For the rest of the SaaS industry, the message is clear: adapt to an interface-free world, or become the invisible, low-margin pipes of the AI revolution.