CrowdStrike is leaning hard into the idea that the next real breakthroughs in cybersecurity won’t come from incremental updates to legacy platforms, but from smaller, faster teams building cloud-native and AI-first defenses from day one. The company just announced the 35 startups selected for the third edition of its Cybersecurity Startup Accelerator, run in close partnership with Amazon Web Services and NVIDIA through the NVIDIA Inception program, and the tone here matters: this isn’t framed as a branding exercise, but as a deliberate attempt to seed the next generation of security architecture. Out of hundreds of global applicants, this cohort was picked for depth of innovation, market relevance, and team quality — the unglamorous but crucial stuff that tends to separate flashy demos from companies that actually survive contact with real attackers.
The eight-week program, running from now through March 3, 2026, is positioned as frictionless for founders: no equity taken, no fees, just access. That access includes hands-on mentorship, deep technical guidance, funding pathways, and go-to-market support, plus something that’s harder to quantify but often more valuable — direct exposure to experienced security operators who have lived through breaches, response chaos, and scale pain. CrowdStrike’s Daniel Bernard described the accelerator as a launchpad for AI-driven security innovators, and the subtext is clear: AI isn’t a feature anymore, it’s the operating assumption. The startups being pulled in are overwhelmingly cloud- and identity-first, designed to operate at the speed and scale of AI-amplified adversaries rather than chasing them after the fact.
AWS’s involvement gives the program its industrial backbone. Chris Grusz, who oversees technology partnerships at AWS, emphasized how these startups are pushing the boundaries of AI-driven security, and that rings true when you look at how much modern security depends on elastic compute, global telemetry, and fast iteration. Meanwhile NVIDIA’s role is increasingly explicit. Bartley Richardson, leading agentic AI and cybersecurity engineering at NVIDIA, framed the accelerator around agentic AI — systems that don’t just analyze threats but act, adapt, and coordinate at machine speed. That’s a quiet but important shift away from dashboards and alerts toward autonomous or semi-autonomous defense, the kind that security teams secretly hope for at 3 a.m. during an incident.
The program will culminate in a final pitch day during the RSA Conference in San Francisco on March 24, 2026, where five finalists will present to an expert panel. One innovation award winner may receive investment from the CrowdStrike Falcon® Fund, which adds a real financial gravity to what might otherwise feel like an industry showcase. RSA, with all its noise and spectacle, remains a meaningful proving ground — a place where startups can test whether their ideas resonate beyond slide decks and into actual buying conversations.
This year’s cohort, excluding stealth companies, reads like a cross-section of where early-stage security thinking is heading: Above Security, Aira Security, Artemis, Astelia, Averlon, Capsule Security, Dash Security, Drift Security, Dux Security, Evoke Security, Fabrix, Fortyx Security, Geordie AI, Haleum AI, Hush Security, Huskeys, Jazz Security, Mars Security, Mate Security, NANO Corp, Nebari, Nimble Security, Opti, Pluto Security, QIZ Security, Raven, Sevii, Simbian AI, SurePath AI, Synqly, Tika Security, VisionHeight, Vivid Security, and Zepo Intelligence. Even the naming tells a story — lots of motion, autonomy, sharp edges, and AI signaling — but the real test will be whether these companies can translate technical promise into tools that security teams actually trust and deploy. CrowdStrike, AWS, and NVIDIA are clearly betting that at least a few of them will, and that those few might quietly reshape how cloud security works over the next decade.
Leave a Reply