Artificial Intelligence & Machine Learning,
Next-Generation Technologies & Secure Development
Acquisitions Add AI Testing and Agent Monitoring to Falcon
In a significant business move, CrowdStrike has announced its decision to acquire two innovative security startups specializing in artificial intelligence (AI). This strategic acquisition is designed to enhance its existing capabilities by introducing tools that assess AI systems for potential vulnerabilities prior to deployment, as well as mechanisms to monitor agent behavior during their operational phase.
The cybersecurity powerhouse will procure Fiddler AI, a company known for its expertise in enterprise AI agent observability software, and Dreadnode, which specializes in automated AI security testing. This dual acquisition aims to broaden the functionality of CrowdStrike’s Falcon platform, extending its coverage across multiple phases of AI deployment, thus reinforcing its commitment to comprehensive cybersecurity solutions.
According to reports from The Wall Street Journal, the financial agreement for these acquisitions amounts to approximately $495 million, with $400 million allocated for Fiddler AI and $95 million for Dreadnode. CrowdStrike has indicated that the majority of the payment will be settled in cash, though a portion will incorporate stock that is subject to vesting conditions. The transactions are anticipated to be finalized by January 31, 2027.
The acquisition of Fiddler AI will equip CrowdStrike with advanced software that not only evaluates the efficacy of AI agents before they are deployed but also provides a robust framework for tracking their decisions and actions once they are active. The technology is set to be marketed under the name Falcon Observe, with availability expected to commence within the month.
George Kurtz, the CEO and founder of CrowdStrike, emphasized the importance of transparency in agent operations. He noted, “Every company is deploying agents and asking the same thing: what are agents actually doing? When a regulator or your board asks what an agent did, saying ‘the model did it’ is not an answer.” This highlights the pressing need for organizations to have clear insights into the decision-making processes of AI agents in real time.
On the other hand, Dreadnode’s technology focuses on preemptively identifying security vulnerabilities within models, applications, agents, and related components before they are put into production. Their platform is uniquely capable of assessing AI systems against over 100 potential attack vectors, including sophisticated strategies such as jailbreaks and prompt injection attacks. Findings from these assessments are prioritized based on severity and aligned with established frameworks, including the OWASP Top 10 for large language models and the National Institute of Standards and Technology’s AI Risk Management Framework.
Moreover, CrowdStrike plans to integrate the testing outcomes from Dreadnode with Falcon Guardian, its existing runtime AI security solution. This integration aims to inform and enhance its protection policies, thereby reinforcing the security architecture of its offerings. The deployment of Dreadnode’s automated AI security testing tools is projected to be available between now and February 2027.
In a statement, Brad Palm, the CEO of Dreadnode, stressed the challenges traditional testing methods face in identifying failures in AI systems. He remarked, “AI fails in ways traditional testing was never designed to catch, and the people best equipped to find those failures think like attackers.” This assertion underscores the critical need for adaptive security measures that can keep pace with the evolving landscape of AI technology.
This acquisition marks the latest phase in CrowdStrike’s ongoing strategy to address AI-related security risks. Earlier in the year, CrowdStrike also agreed to acquire SGNL, enhancing its continuous identity technology and reinforcing access management frameworks. In addition, the purchase of Seraphic further bolstered visibility within browser sessions involving AI applications. Together, these acquisitions reflect a concerted effort to amalgamate identity-based access controls with activity oversight in dynamic environments.