Data + AI Summit Sets the Stage for Databricks’ Next Chapter
Databricks’ latest Data + AI Summit is the annual gathering where the company announces major product releases and strategic moves that define its roadmap for data platforms, enterprise AI development, and next‑generation analytics. This year’s event brings more than 30,000 in‑person attendees to the Moscone Center, with tens of thousands more joining online from over 150 countries, making it the world’s largest data and AI summit. The keynote program features Databricks co‑founders Ali Ghodsi, Matei Zaharia, Arsalan Tavakoli‑Shiraji, and Reynold Xin sharing new product updates, live demos, and customer stories focused on real‑world AI. They are joined by guest speakers including Satya Nadella of Microsoft and Greg Brockman of OpenAI, underscoring the summit’s influence across the AI ecosystem. Alongside more than 800 breakout sessions, Databricks used the stage to spotlight two headline DAIS 2026 announcements: the expanded Agent Bricks platform and the Panther acquisition.
Agent Bricks Evolves into a Comprehensive Agent Platform
Agent Bricks is Databricks’ agent platform designed to help developers build high‑quality AI agents with the model choice, context, and control needed for production use. First introduced at a previous Data + AI Summit, it has already powered over 100,000 agents and now processes more than one quadrillion tokens per year. Customers such as AstraZeneca, 7‑Eleven, Fox Corporation, and Block have shipped agents built on the platform, validating its appeal for enterprise AI development. At DAIS 2026, Databricks presented Agent Bricks as an end‑to‑end agent platform that connects data with AI so teams can focus on agent logic instead of infrastructure. It addresses what Databricks calls the “missing 99%” of agent work: token capacity, deployment, security, evaluation, monitoring, context, and sharing. By embedding these capabilities into the Lakehouse, Agent Bricks aims to make large‑scale agentic systems manageable and governed by default.
Choice, Context, and Control: Inside the Agent Bricks Platform
The expanded Agent Bricks platform centers on three challenges Databricks sees as critical for modern agents: choice, context, and control. On choice, Agent Bricks supports a wide range of proprietary and open‑source models in a single platform, including offerings from OpenAI, Anthropic, Gemini, Qwen, and new support for Kimi, along with a partnership that brings Grok models natively into Databricks. Databricks also highlighted its work on custom models through prompt optimization, fine‑tuning, and reinforcement learning, including data‑specialist agents competitive with leading frontier models. For context, Agent Bricks connects agents to data wherever it lives so they can reason over fragmented enterprise information. For control, it introduces secure sandboxes, agent memory, token capacity management, and governed deployment to reduce risks like prompt injection, data leakage, or runaway costs. According to Databricks, “developers were stuck building infrastructure, not agents,” and Agent Bricks is meant to change that.
Databricks Panther Acquisition Expands Security Lakehouse Vision
Alongside its agent platform news, Databricks announced the Databricks Panther acquisition, agreeing to acquire Panther, an AI security operations center (SOC) platform. Panther is built for a security lakehouse model that aims to replace legacy SIEM tools with an agentic approach to detection and response. The product brings more than 100 out‑of‑the‑box data integrations, detection‑as‑code features, and agent‑driven SOC workflows that automate threat investigations. Trusted by security teams at companies including Anthropic, Panther focuses on helping organizations detect more threats, unify security data, and investigate every alert with swarms of agents. Databricks positions this as a response to AI‑driven attacks, where adversaries now use agents to probe cloud, SaaS, and AI systems faster than manual defenses can respond. The deal is Databricks’ third security acquisition and deepens its investment in building a unified, AI‑ready security stack on Lakehouse.
Full‑Stack Enterprise AI: From Lakehouse to Agents and Security
Taken together, the Agent Bricks platform update and the Panther acquisition signal Databricks’ intention to operate as a full‑stack platform for enterprise AI development and deployment. Agent Bricks gives developers a managed environment to build, evaluate, and scale agents over governed data, while Panther extends that agentic approach to security operations running on a unified security lakehouse. With DAIS 2026 announcements spanning application agents, SOC workflows, and data governance, Databricks is aligning its product strategy around agents that both consume and generate data within the Lakehouse boundary. For customers, the message is that Databricks wants to be the place where AI agents are designed, monitored, and secured, across both business and security use cases. As the Data + AI Summit continues with 800‑plus breakout sessions on engineering, analytics, agents, and AI, these moves frame how Databricks sees the next era of data and AI platforms.
