Databricks buys Antimatter and SiftD.ai on May 18 2026 for AI security

Databricks has added two new startups to its team. This is part of their plan to make AI safer and faster for all users compared to last year.
By Newsroom | Technology, Business |

Databricks, a company that positions itself as a leader in data and AI solutions, has recently acquired two startups: Antimatter and SiftD.ai. The move appears intended to strengthen its new AI security product. The acquisition of Antimatter, which closed last year, was only disclosed recently. SiftD.ai’s deal, by contrast, was rapidly finalized this past Monday. Both companies’ employees are said to have joined Databricks.

SiftD.ai, a relatively new entity that launched its product in November, offered an interactive notebook designed for collaborative work between humans and AI agents. Antimatter’s acquisition likely brings valuable intellectual property to Databricks. These moves follow Databricks' broader strategy of unifying data, analytics, and AI, offering a serverless data warehousing solution built on open lake data with integrated governance and AI capabilities. The platform aims to facilitate the creation of AI agents that continuously refine their performance based on user data, promising faster workflows and deeper insights.

Read More: DeepSeek API 75 percent discount becomes permanent on 23 May 2026

Platform's Foundation and Capabilities

The Databricks platform emphasizes a unified workspace, allowing data scientists, engineers, and analysts to collaborate. It provides tools for data preparation, including cleaning and transformation, as well as real-time data processing for immediate action on incoming streams. For structured data, it leverages Spark SQL, enabling the creation of tables using SQL DDL and seamless integration of SQL queries with Spark programs, supporting various join operations.

Data discovery is also a focus, with Unity Catalog allowing for the tagging and documentation of data assets, alongside a search interface to aid data consumers. The company highlights its serverless Postgres database, integrated with its lakehouse architecture, as being built for the AI era. This approach aims to build and run applications, agents, and AI models directly on organizational data, positioning itself as a "data-aware AI partner."

Read More: Meta Sends 7,000 Staff to AI Jobs Starting May 2026

Company's Trajectory

Databricks has been actively involved in the data and AI landscape, with numerous repositories on platforms like GitHub indicating ongoing development and contributions. The company's mission is stated as helping data teams tackle complex challenges using data and AI.

Frequently Asked Questions

Q: Why did Databricks buy Antimatter and SiftD.ai in May 2026?
Databricks bought these two startups to improve the security of their AI products. By adding these teams, they want to help companies build safer and faster AI tools using their own data.
Q: What does the SiftD.ai deal mean for Databricks users?
The deal for SiftD.ai, which finished on Monday, May 18, 2026, gives users a new tool for humans and AI to work together. This helps teams finish their tasks faster by using interactive notebooks.
Q: How does the Antimatter acquisition help Databricks customers?
The Antimatter deal brings new technology to Databricks that focuses on keeping data safe while using AI. This helps businesses protect their private information when they train their AI models.
Q: What is the main goal of Databricks for its AI platform in 2026?
Databricks wants to be a partner for companies that use AI. They plan to do this by combining data, analytics, and AI tools into one place so teams can build AI agents that learn from their own data.