NewsStocksDatabricks' Ali Ghodsi Says Most Companies Don't Need Smarter AI Models

Databricks' Ali Ghodsi Says Most Companies Don't Need Smarter AI Models

Author: CryptoBriefing·

Key Takeaways

  • Ali Ghodsi contends that AI model intelligence is not the limiting factor for enterprise adoption; the missing element is business-specific context such as proprietary data, workflows, and institutional knowledge.
  • Polling during his 2026 speaking engagements found only about 10% of respondents believe artificial general intelligence has arrived, while roughly 90% rate the AI models they use daily as smarter than most of their colleagues.
  • He advised companies to construct their own enterprise context and ontology rather than buy off-the-shelf AI solutions noting Databricks' Genie platform is designed to capture the data, decisions, and workflows AI systems need for real automation.
  • Even if AI advancement paused entirely, Ghodsi said most companies could still realize years of productivity gains simply by putting their existing data in order.
  • He estimates full enterprise AI adoption could take up to a decade, even for technology-centric firms, because it requires reengineering processes, roles, incentives, and culture.
Databricks' Ali Ghodsi Says Most Companies Don't Need Smarter AI Models

Databricks co-founder and CEO Ali Ghodsi has a message for executives pouring money into the latest frontier AI models: the technology is not the bottleneck. In his view, the problem for most companies is not that AI lacks intelligence, but that it has not been given anything useful about the business it is meant to serve.

Speaking in an appearance on Bloomberg Tech, Ghodsi argued that the current generation of AI models already exceeds what most enterprises can practically use. The gap between what these systems can do and what companies actually get out of them, he said, comes down to a single factor: context.

The 90% paradox

Ghodsi cited a striking disconnect he has observed while polling audiences during his 2026 speaking engagements. Only about 10% of respondents believe artificial general intelligence has arrived. Yet roughly 90% say the AI models they use daily are already smarter than most of their colleagues.

“We don’t need AI to get smarter,” Ghodsi stated. “It just is lacking context.”

The context he described is what he calls “enterprise context”: the proprietary data, institutional knowledge, decision-making workflows, and business logic that make a company’s operations unique. In other words, the models are general-purpose; the missing ingredient is the business-specific knowledge they have no way of knowing on their own.

The data infrastructure argument

Databricks builds its business helping enterprises organize, manage, and activate their data. Its customer roster includes AT&T, Rivian, Adidas, Mercedes-Benz, Unilever, Virgin, and Bayer.

Ghodsi emphasized that companies need to construct their own “enterprise context” and “ontology” — a structured map of how a company’s data, entities, and processes relate to one another — rather than assuming they can simply purchase a turnkey AI solution. In his framing, the decisive AI investment is not access to a smarter model, but the internal groundwork that lets a model act on how the business actually runs. Databricks has been developing tools designed for exactly this purpose. Its Genie platform, for instance, is positioned as a way to capture the enterprise data, decisions, and workflows that AI systems need in order to move from impressive demos to actual automation.

He added that even if AI advancement paused entirely today, most companies would still have years of productivity gains available simply by putting their existing data in order.

A decade of organizational surgery

Ghodsi’s estimate for achieving full enterprise AI adoption is potentially a decade, even for technology-centric firms. That timeline reflects the reality that integrating AI into a business is not a simple software installation. It is an organizational reengineering project that touches processes, roles, incentives, and culture.

For enterprises weighing that argument, the near-term question is not which model to license, but how quickly their data can be organized, connected, and made usable by AI systems. How companies split AI spending between model access and internal data readiness will be one indicator of whether Ghodsi’s view gains ground.

As of September 2026, Ghodsi has continued to press this message, arguing that enterprises should focus on practical AI challenges rather than existential risks.

Source: CryptoBriefing