Databricks strikes deal to buy Panther Labs in cybersecurity push (updated)

Cisco Systems

16/06/2026 16:38

Panther was valued at $1.4 billion after a $120 million Series B round in 2021
Purchase will be Databricks' third cybersecurity acquisition
Databricks aims to compete with security incumbents such as Splunk and Crowdstrike
Adds customer detail in paragraph 4, deal history in paragraph 7-9
SAN FRANCISCO, June 16 (Reuters) - Databricks said on Tuesday it had agreed to buy startup Panther Labs, as the U.S. data analytics provider pushes deeper into the cybersecurity business.
The acquisition, Databricks' third in this area, furthers its aim to compete with security management incumbents like CrowdStrike CRWD.O and Cisco's CSCO.O Splunk.
Databricks declined to comment on the deal terms.
Panther, valued at $1.4 billion after raising $120 million in Series B funding in 2021, gathers data sources and key ingredients for security in one place. That way, AI programs that act with limited human intervention, known as agents, can respond to a rising flood of threats in the AI age. Anthropic is a customer, the companies have said.
Databricks CEO Ali Ghodsi said in an interview that AI had drastically shrunk how much time attackers need to exploit software vulnerabilities, and old ways of managing security information and alerts were "dead." 
"If they're going to attack you with agents, you have to defend with agents," Ghodsi told Reuters, during the company's Data + AI Summit in San Francisco this week. "You have to fight fire with fire."
According to Ghodsi, he first met Panther CEO Jack Naglieri in 2021, had dinner with him and pitched an acquisition years ago without success.
"He wanted to continue going alone," Ghodsi said. "That was a good idea for him because his valuation has been going up since then and he's built an awesome platform."
At $134 billion, Databricks is one of the world's most valuable privately held companies. It unveiled its security ambitions in March with a product called Lakewatch.

 (Reporting by Jeffrey Dastin in San Francisco; Editing by Chizu Nomiyama, Franklin Paul and Mark Porter)
 ((Jeffrey.Dastin@thomsonreuters.com; +1 424 434 7548))