Databricks launches background AI agents to automate office tasks

The move toward agentic AI

Databricks recently unveiled a new suite of agentic AI tools designed to handle the heavy lifting of data organization for office workers. Unlike previous models that require a user to initiate a specific request, these agents are designed to run continuously in the background. They analyze and organize data scattered across a business to help employees perform their jobs more efficiently.

The company describes these tools as being 'always on' to provide a constant layer of assistance. By shadowing a worker throughout the day, the AI can proactively offer suggestions or pull up specific information based on the current context of the user's work. This approach is intended to transform how employees interact with their daily tasks and data sets.

CEO and cofounder Ali Ghodsi stated that he uses the tool every hour of every day. He claimed that the technology has personally made him a 10x better CEO over the last year. This personal endorsement highlights the company's goal of making AI a primary driver of individual productivity.

How the Genie and Ontology work

The updated version of the Genie assistant serves as a proactive companion for the knowledge worker. It monitors the flow of work to identify what information is needed next, effectively acting as a digital assistant that anticipates needs. This allows workers to spend less time searching for files or manually compiling reports.

Complementing Genie is a separate feature called 'Ontology.' This tool scans the entire data landscape of an organization, including management reporting lines and various departmental projects. By mapping out these connections, the system allows any individual within the company to quickly access specific information or run complex analyses across different teams.

The goal of these combined features is to create a cohesive map of a company's internal knowledge. By organizing this data automatically, Databricks hopes to make it easier for both large and small companies to leverage their existing information. The technology aims to bridge the gap between raw data and actionable insights for the average office employee.

A broader industry shift

The launch of these tools follows a similar move by Anthropic, which recently released its own productivity suite. Anthropic introduced an AI-powered spreadsheet application to compete with established software like Microsoft Excel and Google Sheets. They also launched a feature that allows their Claude AI to join video calls to listen and start working while people are still speaking.

These developments suggest that the AI industry is moving beyond the initial success in software programming. While programmers have already integrated AI into their workflows, the technology is still in an experimental phase for many other business functions. Companies are now looking for clearer returns on investment in areas like marketing, HR, and general administration.

The industry is currently making a full court press to broaden its base of paying customers. By targeting the knowledge worker, AI firms hope to capture a larger share of the global economy. The focus is on moving from specialized tools to general-purpose productivity assistants that can handle a wide variety of corporate tasks.

Economic stakes and IPOs

The push toward productivity is tied to significant economic goals and upcoming financial milestones for major AI players. Anthropic recently noted in its prospectus that knowledge work represents more than $40 trillion in global economic value for 2025. They believe that using frontier AI to make these workers more productive is a primary opportunity for growth.

Anthropic is currently preparing for an initial public offering (IPO) expected next week, which could potentially value the company at $1.4 trillion. A successful debut on Wall Street would serve as a major signal for the health and viability of the broader AI industry. Investors are watching closely to see if these productivity tools can translate into massive commercial success.

Meanwhile, Databricks remains a private company but recently raised $5 billion at a $190 billion valuation. Despite some recent distractions in the industry, such as reports regarding Nvidia's leadership, Databricks leadership remains confident. Ghodsi expressed belief that the AI revolution is too large to be derailed by individual leadership changes at the chip manufacturing level.