Dreadnode Deal Would Bring Pre-Deployment AI Testing to Falcon

CrowdStrike announced on November 3, 2026 (CRWD ) that it has signed a definitive agreement to acquire Dreadnode, an automated AI security-testing company. The planned acquisition would add continuous testing of AI models, applications and agents to the Falcon platform, with the aim of identifying weaknesses before those systems enter production.

The consideration is contemplated to be paid predominantly in cash, with a portion in stock subject to vesting conditions. CrowdStrike did not disclose a dollar purchase price. It expects the transaction to close during its fourth quarter of fiscal 2027, subject to customary closing conditions.

Testing AI Before Deployment

The deal addresses a difference between conventional software testing and AI systems. Traditional testing was developed for code that behaves consistently; models and agents are probabilistic and can fail through jailbreaks, prompt injection or manipulation of an agent’s reasoning process. The planned product moves testing into the layer where enterprises build and assess AI, extending CrowdStrike’s AI Detection and Response category into pre-deployment testing.

CrowdStrike CEO and founder George Kurtz framed the acquisition around that visibility gap: “Enterprises are deploying AI faster than they can prove it is secure. A model can be jailbroken, an agent can be manipulated, and most organizations have no way to find those flaws until the AI is already in production.”

The planned workflow allows users to describe tests in plain language. CrowdStrike says Dreadnode runs more than 100 known AI attack strategies, including jailbreaks, prompt injection, multi-turn attacks and agent-specific attacks, against models, agent reasoning loops, applications and MCP servers. A separate scanning function examines AI models, model repositories and components for vulnerabilities before production.

Results are ranked by severity and mapped to the OWASP Top 10 for LLMs, OWASP Agentic Security ASI01-ASI10, MITRE ATLAS and NIST AI RMF. Exportable reports are designed to support go/no-go deployment decisions. Human reviewers can reclassify findings, change severity ratings and document their reasoning, with dashboard metrics recalculating automatically.

Feeding Findings Back Into Protection

The acquisition also connects testing with CrowdStrike’s runtime and research operations. Dreadnode findings are planned to feed Falcon Guardian as policy recommendations, allowing a weakness found before deployment to inform runtime protection. Dreadnode’s testing data will be used to strengthen SafeMind models, while its technology is expected to sharpen how SafeMind measures frontier models’ security and cyber capabilities.

The Dreadnode team will join CrowdStrike’s Cyber Superintelligence Lab, adding AI security researchers to the group. That makes the transaction more than a product acquisition: it also brings the research staff and testing data behind the planned capability into CrowdStrike.

Automated AI Security Testing is targeted for general availability in the fourth quarter of fiscal 2027. With the product and expected closing targeted for the same fiscal quarter, the deal would give CrowdStrike a workflow spanning pre-deployment tests, runtime policy recommendations and model research. Its significance therefore rests not only on adding another testing tool, but on combining those pieces into a single path from discovering an AI weakness to protecting against it.