Kai

Kai is building what it calls an agentic AI security platform: autonomous software agents that ingest telemetry from enterprise networks, cloud environments, endpoints, and existing security tools, then analyze it and take defensive action — flagging or responding to anomalous behavior, vulnerabilities, and indicators of compromise — without requiring a human analyst in the loop for every decision. The framing is explicitly adversarial: as attackers increasingly use AI and automation to scale their operations, Kai’s pitch is that defense needs an AI system operating at comparable speed rather than a human SOC team working through AI-augmented but still manual workflows.

Kai launched from stealth in March 2026 with $125 million in funding led by Evolution Equity Partners, with participation from N47 and other investors — a large round for a company this new, reflecting investor confidence in its founders more than any yet-established track record. It was founded in 2025 by Galina Antova, who previously co-founded industrial cybersecurity company Claroty, and Damiano Bolzoni, who co-founded SecurityMatter (acquired by Forescout). The company is based in San Jose, California.

Because Kai is pre-track-record, the honest evaluation here rests almost entirely on founder pedigree and capital raised rather than demonstrated product outcomes: Antova and Bolzoni both built and sold security companies before, which is a meaningful signal, but there is no public case study, customer name, or independent evaluation yet of how well the agentic security platform actually performs in production. Its differentiation claim — collapsing siloed detection, analysis, and response tools into one integrated agent pipeline — is also the same pitch made by several other well-funded "agentic SOC" entrants, so its durability will depend on execution that has not yet been publicly demonstrated.