Startup gets $75M to secure AI in cloud platforms

Israeli cybersecurity startup Sweet Security has secured $75 million in a Series B funding round to aggressively integrate artificial intelligence security capabilities into its Cloud-Native Application Protection Platform, according to GovInfoSecurity.

The company, led by former Israeli Military Intelligence veterans, plans to leverage its existing runtime infrastructure and eBPF sensors to provide deep behavioral monitoring and guardrails for non-deterministic AI agents operating in cloud environments. Co-founder Eyal Fisher stated that current CNAPP tools are insufficient for the unique risks posed by autonomous AI, creating a significant market need that Sweet is positioned to address.

The firm’s approach involves baselining normal AI agent activity to detect and flag anomalous behaviors, such as an agent suddenly accessing an unfamiliar sensitive directory. Fisher emphasized that AI security must remain the responsibility of the existing security team and SOC to avoid organizational gaps. With this new capital, Sweet Security intends to triple its go-to-market efforts, focusing on expanding proof-of-concept deployments and winning more enterprise customers.