CrowdStrike Acquires Fiddler and Dreadnode as SOCs and Cities Scale AI

Enterprise security vendors and municipal technology operators are re-architecting their software stacks around autonomous agent oversight, continuous red-teaming, and shared GPU infrastructure. From pre-deployment testing and runtime observability to security operations center staffing and urban digital twins, organizations are replacing manual queue triage with engineering controls designed to trace, validate, and halt autonomous models in production.

CrowdStrike Signs Deals for Fiddler AI and Dreadnode

According to CyberScoop, CrowdStrike announced Monday at its Fal.Con Europe conference in Barcelona that it has signed agreements to acquire two artificial intelligence security companies, Fiddler AI and Dreadnode. While CrowdStrike did not disclose financial terms, a Wall Street Journal report listed the purchase price for Fiddler AI at $400 million and Dreadnode at $95 million. Both transactions are expected to be paid mostly in cash alongside performance-conditioned company stock and are scheduled to close in CrowdStrike's fourth quarter of fiscal 2027, when both products are also expected to reach general availability.

Fiddler AI specializes in AI agent observability and explainability and will be integrated into the Falcon platform as Falcon Observe. The software evaluates agents prior to production, traces decisions and reasoning across agent systems, flags drifting behavior and fabricated answers, and generates audit records across common frameworks including LangGraph, Amazon Bedrock, and Google ADK. CrowdStrike Chief Executive and founder George Kurtz described Fiddler as "the flight recorder for every agent," noting that telling a regulator or board "the model did it" is not an acceptable answer.

Dreadnode operates earlier in the lifecycle by running automated security testing across more than 100 known attack methods—including prompt injection and jailbreaks—against models and agent systems, ranking weaknesses by severity and keeping them aligned to widely used security standards such as the OWASP Top 10 for large language models and the NIST AI Risk Management Framework. Dreadnode Chief Executive Brad Palm noted that the company was founded on the premise that testers who think like attackers are best equipped to uncover failures. Alongside new Falcon Guardian runtime protections, Kurtz also introduced an external runtime control called the AI Kill Switch, designed to let human operators stop any autonomous agent immediately.

Gartner Projects 50% More SOC Engineers Than Analysts by 2028

As automated agents absorb front-line security queues, enterprise staffing models are shifting toward systems engineering. As reported by BleepingComputer, a 2026 Gartner report titled The Roles Required for the AI-Enabled Security Operations Center (SOC) by analyst Pete Shoard predicts that security operations teams will have 50% more engineers than analysts by 2028. Because AI SOC agents significantly reduce alert-handling work for traditional detection cases, Gartner advises that human analysts transition from alert triage to end-to-end case ownership and response option communication.

Gartner states that the SOC will not shrink by 2030; instead, personnel previously assigned to alert volume are redirected into detection engineering, threat hunting, offensive testing, and continuous threat exposure management (CTEM). In Prophet Security's State of AI in Security Operations 2026 survey of more than 250 CISOs, SOC leaders, and practitioners, 57% expect their team size to stay the same and 9% expect growth. Furthermore, 49% of teams that conduct weekly hunting found malicious activity that automated detections missed, compared with 8% among teams reporting the lowest hunting frequency. Gartner urges leaders to replace alert-volume metrics with case accuracy, cycle time, and decision confidence, while expanding detection engineering to encompass prompt design, workflow backtesting, AI output validation, and version-controlled rollback, supported by ISO 2859-1 sampling audits.

NVIDIA and Municipal Partners Expand Physical and Agentic AI

Public-sector agencies are similarly deploying agentic workflows and simulation models atop shared computing infrastructure. According to the NVIDIA Blog, NVIDIA and its global partners are expanding physical and agentic AI services ahead of the Smart City Expo World Congress running November 3 to 5 in Barcelona, where NVIDIA Vice President of EMEA Paolo Guglielmini and EMEA smart city director Charbel Aoun are scheduled to speak.

In the San Francisco Bay Area, LiveX is combining NVIDIA Cosmos world models, Nemotron open models, and Omniverse and Metropolis libraries on NVIDIA RTX PRO GPUs to deploy interactive AI hosts across airports, hospitals, and public spaces. In Spain, 1MillionBot pairs open-source ALIA models trained at the Barcelona Supercomputing Center with NVIDIA Nemotron across more than 150 deployments, including Madrid Digital and Administració Oberta de Catalunya, to manage municipal procedures and service inquiries.

In Wrocław, Poland, K2K is building shared municipal AI infrastructure using Cosmos world reasoning models and the NVIDIA Metropolis Blueprint for video search and summarization (VSS). In Malaysia, ITMAX is scaling its SamurAI V2 vision agents on NVIDIA RTX PRO 6000 GPUs to support more than 17,000 cameras by 2027 while integrating the open-source TeraSim traffic platform, cutting peak-hour corridor travel times by up to 26%. Meanwhile, at Dangote's Lekki refinery in Nigeria, Honeywell Technologies' Forge Plant Cognition runs Nemotron agents on Dell PowerEdge servers, Cires Technologies is expanding a sovereign AI factory at Morocco's Tanger Med port, and in Taiwan, Linker Vision and Dutch architecture firm MVRDV are using Omniverse and Nemotron to simulate 15-year outcomes of an urban forest canopy in 10 minutes instead of 2 weeks—accelerating design iterations to less than an hour.