Dawnguard and Cyata Step Out of Stealth to Tackle AI Security at the Source

Two new startups, Dawnguard and Cyata, have emerged from stealth with a shared goal: to rethink how cybersecurity adapts to the rise of AI-native systems. Dawnguard, based in Amsterdam, is building a design-first security platform that embeds protection into the architecture of cloud environments. Tel Aviv-based Cyata is focused on controlling the new AI workforce, autonomous agents that interact with enterprise systems without traditional identity guardrails.

Dawnguard raised $3 million in pre-seed funding to grow its engineering team and expand enterprise integrations. The platform is designed for security architects, DevOps engineers, and cloud teams who want to ensure infrastructure is secure and compliant from the design phase onward. Instead of reacting to threats after deployment, Dawnguard allows teams to generate validated Infrastructure as Code and enforce posture continuously, creating alignment between security intent and implementation.

Cyata, meanwhile, raised $8.5 million in seed funding led by TLV Partners. Its platform addresses the emerging risks of AI agents - digital entities that act autonomously, write code, access data, and make decisions without human oversight. These agents don’t fit within the old identity and access management models. Cyata brings visibility and control to this layer through automated discovery, forensic observability, and granular access enforcement, with features built specifically for AI identity workflows.

As AI agents move from experiments to essential infrastructure, platforms like Dawnguard and Cyata are pushing the industry to treat security as part of the build, not just the patch. This is where the trust gap gets closed.