CrowdStrike, a dominant force in the global cybersecurity landscape, announced on Tuesday a significant expansion of its artificial intelligence security capabilities through the acquisition of two specialized startups: AI agent observability firm Fiddler AI and offensive AI security testing provider Dreadnode. The dual acquisitions, reportedly valued at a combined $495 million, signal a strategic pivot by the cybersecurity giant to address the burgeoning security risks inherent in the rapid deployment of autonomous AI agents within corporate environments.
While CrowdStrike has maintained a degree of confidentiality regarding the specific financial breakdowns of these deals, reports from The Wall Street Journal—which surfaced ahead of the official company disclosure—indicate that the company is investing approximately $400 million to acquire Fiddler AI, while allocating roughly $95 million for the purchase of Dreadnode. According to the official statement released by the company, these acquisitions will be funded primarily through cash, with a smaller portion of the consideration composed of company stock subject to standard vesting conditions. The transactions are currently slated to finalize during the fourth quarter of CrowdStrike’s fiscal year 2027, which concludes on January 31, 2027.
Scaling AI Observability with Fiddler AI
Founded in 2018 and headquartered in Palo Alto, California, Fiddler AI has spent the last several years positioning itself as a leader in the complex field of AI observability and explainability. Having secured $100 million in total venture funding—including a robust $30 million Series C round announced earlier this year in January—the startup has developed a reputation for providing transparency into the “black box” nature of artificial intelligence systems.
For CrowdStrike, integrating Fiddler’s technology is a strategic move to provide its enterprise customers with a deeper understanding of how their AI agents function in real-world scenarios. The core value proposition of Fiddler lies in its ability to monitor AI agents, track their reasoning processes, and explain the decisions they reach. By incorporating Fiddler’s suite into the broader Falcon ecosystem, CrowdStrike intends to offer pre-deployment evaluation tools, allowing security teams to trace the logic of tool calls and identify potential behavioral drift or “hallucinations” before they impact production environments.
Furthermore, the technology addresses the increasing pressure on enterprises to meet regulatory and internal compliance standards. By logging agent activity for audit purposes and providing granular tracking of token usage and associated operational costs, the acquisition ensures that CrowdStrike’s clients can maintain oversight of their AI expenditures and security posture simultaneously. CrowdStrike has indicated that it plans to market these new capabilities under the "Falcon Observe" banner, aiming to make it a cornerstone of its AI-driven security operations.
Strengthening Offensive Defenses with Dreadnode
While Fiddler AI focuses on observability, the acquisition of the 2023-founded startup Dreadnode provides CrowdStrike with a formidable offensive security component. Despite being a relatively young entity, Dreadnode has quickly gained industry attention for its specialized platform designed to stress-test AI systems. The startup recently secured $14 million in Series A funding in early 2025, underscoring the market’s high valuation of offensive AI security.
Dreadnode’s platform is designed for ease of use, allowing users to define testing parameters through plain language prompts. Once a test is initiated, the platform executes a battery of over 100 known AI attack strategies, ranging from sophisticated jailbreaks and prompt injection techniques to complex multi-turn and agent-specific exploits. These tests are comprehensive, targeting the underlying AI models, agent reasoning loops, various applications, and Model Context Protocol (MCP) servers.
Crucially, the findings from these tests are not just raw data; they are ranked by severity and mapped against industry-standard frameworks such as the OWASP Top 10 for LLMs, the MITRE ATLAS framework, and the NIST AI Risk Management Framework. These insights are then integrated directly into CrowdStrike’s "Falcon Guardian" AI detection and response product, which translates the findings into actionable policy recommendations for administrators. Under the terms of the deal, the Dreadnode team, led by CEO Brad Palm, will be integrated into CrowdStrike’s Cyber Superintelligence Lab. The company expects the Dreadnode-based testing capabilities to become generally available to customers by the fourth quarter of fiscal 2027.
A New Era of AI Governance: The "Kill Switch"
The dual acquisition announcement was accompanied by the unveiling of a new safety feature within the Falcon Guardian platform: the "AI Kill Switch." This development highlights CrowdStrike’s commitment to not only monitoring and testing AI but also providing a direct mechanism for incident response.
The AI Kill Switch is designed to empower authorized administrators with the ability to instantly halt a misbehaving or compromised AI agent. Once activated, the feature not only stops the agent’s execution but also prevents it from automatically restarting and severs its network access, effectively neutralizing a threat before it can propagate through a corporate network. This functionality serves as a vital safeguard for organizations that are increasingly reliant on autonomous agents to perform sensitive tasks, providing a "break-glass" protocol that is essential in the high-stakes world of enterprise security.
Sustaining an Aggressive M&A Strategy
The acquisitions of Fiddler AI and Dreadnode are part of a broader, aggressive acquisition strategy that CrowdStrike has pursued over the past year to cement its position as a holistic security provider. The company has been systematically building its capabilities in identity security, browser security, and AI detection.
Earlier in 2026, the company announced its intent to acquire the identity security firm SGNL for $740 million, followed shortly by the acquisition of browser security company Seraphic for $420 million. Additionally, in September 2025, the firm reached an agreement to acquire the AI security startup Pangea, a move that set the stage for the launch of its current AI detection and response (AIDR) initiatives.
As the cybersecurity industry continues to navigate the dual challenges of protecting against AI-powered threats while simultaneously securing the AI systems that enterprises are rapidly adopting, these high-profile acquisitions reflect a shifting paradigm. CrowdStrike is clearly betting that the future of cybersecurity is intrinsically linked to the governance and protection of the AI agents that are becoming the new engines of corporate productivity. By combining observability, offensive testing, and direct response capabilities, the firm is attempting to create a defensive moat around the AI lifecycle, ensuring that as organizations scale their use of LLMs and autonomous agents, they do so with a comprehensive layer of security that matches the speed and complexity of the technology itself. With these latest investments, CrowdStrike continues to signal to the market that it is prioritizing the maturation of the AI security stack as a top-tier operational imperative for the years ahead.






