CrowdStrike's AI Security Buying Spree

Two acquisitions and a new emergency control, announced together, sketch out what CrowdStrike thinks enterprise AI security looks like: watch what an agent does, test it before deployment, and keep a switch that can stop it when something goes wrong. The company said on Tuesday it had agreed to acquire AI agent observability firm Fiddler AI and AI security testing company Dreadnode. The Wall Street Journal, which reported the deals ahead of the official announcements, put the combined price at roughly $495 million.

Two targets, one thesis

Fiddler and Dreadnode sit on opposite sides of the same problem. Fiddler watches AI agents in production and explains why they made the decisions they did. Dreadnode attacks AI systems on purpose, running known attack strategies against models, agents and the servers they connect to. One is defensive telemetry; the other is an adversary simulation that feeds policy back into CrowdStrike's detection stack.

The two companies are at very different stages. Fiddler was founded in 2018 and is based in Palo Alto, California. In January it announced a $30 million Series C round that brought its total funding to $100 million. Dreadnode was founded in 2023 and raised $14 million in Series A funding in early 2025. Its team is led by CEO Brad Palm and will join CrowdStrike's Cyber Superintelligence Lab.

What the price tag reportedly covers

CrowdStrike has not disclosed the financial terms of either deal. The breakdown comes from the Journal's reporting: roughly $400 million for Fiddler and about $95 million for Dreadnode. CrowdStrike said both purchases will be paid mostly in cash, with a portion in stock subject to vesting conditions.

The deals are expected to close in the fourth quarter of CrowdStrike's fiscal year 2027, which ends on January 31, 2027. Dreadnode-based testing capabilities are expected to become generally available in that same quarter.

Inside Fiddler's observability stack

CrowdStrike says Fiddler's technology will add pre-deployment evaluation of agents, tracing of their reasoning and tool calls, and detection of behavioral drift and hallucinations once those agents are running in production. It will also log agent activity for compliance audits and track token usage and spending — a cost-control function that sits unusually close to security monitoring.

CrowdStrike plans to sell the technology as Falcon Observe. The pitch rests on a difficult property of agentic systems: an agent can be working exactly as configured and still produce an outcome nobody intended, because the reasoning path is opaque. Tracing reasoning and tool calls is meant to make that path visible; drift and hallucination detection is meant to catch it when the path changes without anyone shipping a new version.

Dreadnode's offensive playbook

Dreadnode's platform is built around plain-language test descriptions. A user describes what they want to test, and the platform runs more than 100 known AI attack strategies, including jailbreaks, prompt injection, and multi-turn and agent-specific attacks. The targets are AI models, agent reasoning loops, applications and MCP servers.

Findings are ranked by severity and mapped to the OWASP Top 10 for LLMs, MITRE ATLAS and the NIST AI RMF. They are then fed to CrowdStrike's Falcon Guardian AI detection and response product as policy recommendations — closing a loop in which offensive testing directly tunes the defensive configuration.

The kill switch for runaway agents

Alongside the acquisitions, CrowdStrike introduced an AI Kill Switch in Falcon Guardian. It allows authorized administrators to halt a misbehaving AI agent, prevent it from restarting, and cut off its network access. The three actions are distinct: stopping the process, blocking its return, and severing what it can reach.

The feature addresses a scenario security teams have struggled to operationalize — an autonomous process that is doing harm and needs to be contained quickly, without waiting on a change-management cycle or hunting down every host it touches.

The year of buying AI security

These are not CrowdStrike's first acquisitions in this area. In January, it announced plans to buy identity security firm SGNL for $740 million and browser security company Seraphic for a reported $420 million. In September 2025, it agreed to acquire AI security firm Pangea.

Identity, browser, AI detection, agent observability, offensive AI testing. Each purchase adds a different layer of the same enterprise estate, and each was announced as a piece of the Falcon platform rather than a standalone product line.

The numbers behind the deals

• $495 million — combined reported value of the Fiddler and Dreadnode deals, per The Wall Street Journal
• $400 million — reported price for Fiddler
• $95 million — reported price for Dreadnode
• $100 million — Fiddler's total funding to date, including a $30 million Series C announced in January
• $14 million — Dreadnode's Series A funding, raised in early 2025
• January 31, 2027 — end of CrowdStrike's fiscal year 2027, the quarter in which both deals are expected to close
• 100+ — known AI attack strategies Dreadnode's platform runs

Why it matters

For security teams, the practical question is not whether CrowdStrike can bolt two more products onto Falcon, but whether agent observability and adversarial testing become table stakes for anyone deploying autonomous AI. If CrowdStrike's bet is right, the answer could shape procurement: a buyer evaluating AI agents may soon expect tracing, drift detection and a documented kill path as part of the security stack rather than as separate tools.

There is a second, quieter implication. The Kill Switch exists because agents can misbehave in ways that are hard to predict in advance. A vendor shipping an emergency stop alongside observability and offensive testing is, in effect, acknowledging that prevention alone is not enough for systems that reason on their own. Whether customers pay for all three layers — or pick one — is the commercial test the deals will ultimately face.