CrowdStrike is acquiring two AI security companies to expand its products to monitor, test, and cover AI models and agents as enterprises deploy these technologies wholesale. According to The Wall Street Journal, CrowdStrike is paying around $400 million for Fiddler AI and about $95 million for Dreadnode.
Corporate policies and processes currently lag behind internal AI adoption. Across various departments, teams and employees are automating workflows, building custom apps through vibe-coding, and deploying autonomous agents to handle operations, generating new exposure for enterprise IT environments.
"The deals are happening at a time of significant upheaval in the cybersecurity space, especially as companies’ policies and processes lag behind AI adoption within their walls."
This rapid spread of internal systems creates operational blind spots, establishing dedicated AI supervision as an emerging requirement in corporate technology budgets.
Integrating Observability into Falcon
The bulk of CrowdStrike's capital outlay targets observability rather than pre-deployment scanning alone. CrowdStrike is allocating roughly $400 million specifically for Fiddler AI, a startup founded in 2018 that had raised a total of about $99 million to date according to PitchBook, marking a significant liquidity event for its backers and reflecting the high price of established market presence in AI monitoring.
Fiddler AI develops observability software to monitor AI agents inside corporate environments. CrowdStrike will integrate Fiddler AI into its Falcon platform under the name Falcon Observe. The system will evaluate agents prior to rollout, track their operations, and stream that telemetry directly into the broader Falcon architecture, adding a distinct line item to corporate IT budgets for continuous agent monitoring.
To address pre-release vulnerabilities, CrowdStrike is integrating Dreadnode into its Falcon Guardian product. Dreadnode, founded in 2023 with $14 million in funding raised so far, automates security testing designed to identify flaws in AI models, applications, and agents before they go live.
"Dreadnode, meanwhile, will be integrated into the Falcon Guardian product, to which the startup’s software will feed its testing results as security policy recommendations."
CrowdStrike has been developing internal detection and response mechanisms under the Falcon Guardian moniker to help companies track agent activity across endpoints, and Dreadnode's testing output will feed policy recommendations straight into that engine.
Enterprise Defenses and Tooling Gaps
A counter-argument in enterprise IT management suggests that existing endpoint and network security platforms might already provide adequate controls without dedicated acquisitions. However, emerging security architectures are heavily shaped by the conviction that legacy software cannot properly monitor and control AI tools, connectors, or runtime behavior. While CrowdStrike leverages its presence across enterprises and governments, several startups and cybersecurity vendors have also raised dozens of millions in funding to address these exact visibility gaps.
Enterprises rapidly unleash autonomous agents to eliminate manual labor, only to immediately purchase specialized nine-figure software stacks to watch what those agents are doing, reshaping enterprise IT budgets around dedicated AI oversight.
What changed
Enterprises roll out autonomous agents to cut operating expenses, only to instantly generate a mandatory new budget line item to oversee them. Shadow scripts and custom employee apps create operational blind spots that legacy endpoint platforms fail to track during runtime. CrowdStrike tackles this visibility gap through acquisitions, folding specialized testing and agent observability straight into its core Falcon architecture.