CrowdStrike spends nearly $500M on Fiddler AI and Dreadnode to secure AI agents

What happened

CrowdStrike is buying two AI security startups, Fiddler AI and Dreadnode, in deals worth close to half a billion dollars combined. According to figures reported by The Wall Street Journal and cited by TechCrunch, the company is paying around $400 million for Fiddler AI and roughly $95 million for Dreadnode.

The two companies cover different ends of the AI security lifecycle. Fiddler AI, founded in 2018, builds observability software that lets enterprises monitor the AI agents running inside their environments. Dreadnode, founded in 2023, automates security testing aimed at finding flaws in AI models, applications and agents before they go live. Per PitchBook data cited by TechCrunch, Fiddler AI had raised about $99 million in total funding to date, while Dreadnode had raised $14 million.

Where they land in Falcon

Both products are being folded into CrowdStrike's Falcon platform, but at different points. Fiddler AI will become Falcon Observe, where according to TechCrunch it is expected to help customers evaluate agents before deployment, understand and monitor what those agents actually do, and pipe that activity back into the platform as telemetry.

Dreadnode, meanwhile, will be integrated into Falcon Guardian, the detection and response layer CrowdStrike has been building to give companies visibility into what agents are doing on their devices. The startup's automated testing results will feed into that product as security policy recommendations.

Taken together, the pairing suggests CrowdStrike intends to cover the full arc of an agent's life: test it before launch, watch it while it runs, and translate the findings into enforceable policy.

A category forming quickly

The acquisitions land amid what TechCrunch describes as significant upheaval in cybersecurity, with corporate policies and processes lagging behind AI adoption inside companies. Teams and individual employees are experimenting with AI tools, automating workflows, assembling custom applications at speed and deploying large numbers of agents — all of which creates headaches for security teams trying to keep track of what is connecting to what.

CrowdStrike, whose products are widely used across enterprises and governments, is using its existing heft and presence to claim a share of this fast-growing market. As TechCrunch notes, it is not alone: a wave of startups and established security vendors have raised substantial funding on the shared premise that existing security software was not designed to monitor and control AI tools, connectors or usage.

Why it matters

Nearly $500 million for two companies that raised a combined $113 million signals how quickly securing AI models and agents has moved from a niche concern to a category that large vendors will pay a premium to enter. Fiddler AI sold for roughly four times the total capital it had raised, and Dreadnode for close to seven times its own.

The structure of the deals also indicates where the market believes the risk sits. Observability and pre-deployment testing target the moment enterprises are most exposed: when autonomous agents begin acting on systems and data with limited human review. Embedding those controls into a platform already deployed across much of the enterprise and government world could turn AI security into a default feature rather than a specialist purchase.

If agent adoption keeps accelerating, further consolidation looks likely. Incumbent security vendors face a build-or-buy decision on AI-specific tooling, and CrowdStrike has now set a reference price for buying.