CrowdStrike buys identity startup SGNL for $740M to counter AI-driven cyberattacks
CrowdStrike is making a big move into identity security, announcing Thursday that it will acquire startup SGNL in a deal valued at nearly $740 million. The bet reflects a growing concern across enterprise security teams: as artificial intelligence reshapes how attacks are carried out, identity has become the softest entry point.
Investors reacted quickly. CrowdStrike shares slipped nearly 4% following the announcement.
The acquisition brings SGNL’s identity authorization technology into CrowdStrike’s Falcon cloud security platform, adding tighter controls over how both humans and AI agents request access across systems. The company says the deal will help customers evaluate access requests in real time, factoring in context and risk before granting permissions. The transaction is expected to close in the first quarter of CrowdStrike’s 2027 fiscal year.
“This is a massive opportunity for our customers to be able to protect themselves, and a massive opportunity for us to disrupt the identity market,” CEO George Kurtz said in an exclusive interview with CNBC.
CrowdStrike Acquires SGNL in $740M Deal to Strengthen Identity Security as AI Threats Grow
Identity has quietly become one of the most common ways attackers break in. Credentials are stolen, misused, or abused by automated tools that move faster than human defenders. CrowdStrike says its identity security business reached $435 million by the end of the second quarter, turning it into one of the company’s fastest-growing attack surfaces.
AI has only raised the stakes. Enterprises are deploying autonomous agents to manage workflows, infrastructure, and data access, introducing new non-human identities that still require guardrails. Security teams are under pressure to manage this sprawl without slowing operations.
The industry has already seen how damaging identity-based attacks can be. Microsoft faced a wave of breaches last year tied to its SharePoint platform, and AI research company Anthropic disclosed what it described as the first documented AI-led cyberattack in November. Those incidents have pushed identity security higher on boardroom agendas.
The acquisition also carries historical weight. In 2024, CrowdStrike was at the center of a global tech outage triggered by a faulty software update, where a single line of C++ code cascaded across systems worldwide. The incident underscored how tightly coupled modern infrastructure has become—and how quickly failures can spread.
SGNL, based in Palo Alto, raised $30 million in an early funding round in February. Its investors include Cisco Investments and Microsoft’s venture fund. The company was founded in 2021 by Scott Kriz and Erik Gustavson, whose previous startup was acquired by Google in 2017. Both later spent more than four years at Google, working on large-scale systems.
CrowdStrike’s move fits into a broader acquisition streak across the cybersecurity sector. Vendors are racing to assemble broader platforms as customers push to reduce tool sprawl and vendor overlap. Big checks have followed. Palo Alto Networks agreed to acquire Israeli startup CyberArk for $25 billion, while Google bought cloud security company Wiz for $32 billion.
CrowdStrike has taken a similar path. In 2025, the company announced plans to acquire the AI-agent security platform Pangea and the Spanish data startup Onum.
Kurtz says the strategy is straightforward: buy winning teams and technology built for current threats, not yesterday’s problems.
“We want to offer the most value to our customers where they can consolidate on CrowdStrike — fewer vendors, fewer complexities, fewer costs, and with a better outcome of stopping breaches,” he said.
Founded in 2011 by Dmitri Alperovitch, George Kurtz, and Gregg Marston, CrowdStrike built its reputation on cloud-based endpoint protection. Over the years, it has steadily broadened its scope, moving from devices to workloads, data, and now identity. The SGNL acquisition signals where the company thinks the next major security battles will be fought — and who needs to be guarded, human or machine.


