Ent.AI raises $100M seed—one of largest cybersecurity rounds ever—for intent-aware workspace prevention

Ent emerged from stealth on June 16 with $100 million in seed funding—one of the largest seed rounds in cybersecurity history. The round was led by Decibel Partners, with participation from Sequoia Capital, Crosspoint Capital, Craft Ventures, Shield Capital, Felicis, and In-Q-Tel (CIA-affiliated venture fund). Founded by Elias Manousos and Brandon Dixon, the RiskIQ co-founders (acquired by Microsoft for $500M+ in 2021) who also built Microsoft Security Copilot, Ent is already deployed with Global 2000 customers in hospitality, financial services, and defense sectors.

Ent's core bet: endpoint security has spent a decade in a reactive detection loop, but AI-powered attacks demand a new paradigm: on-device AI reasoning for real-time prevention. Instead of waiting for post-breach logs, Ent runs specialized AI models at the endpoint to understand intent—both of human users and AI agents—and intervenes at the moment of decision. The company positions itself as a layer above traditional EDR (endpoint detection and response), solving problems that legacy tools were not architected to handle: governing AI-agent behavior, detecting insider risks that blend into normal workflows, and preventing data loss before exfiltration.

For practitioners: endpoint security is a $7.23B market projected to reach $45.95B by 2034. Ent's timing is deliberate: as organizations deploy autonomous AI agents, traditional behavioral analytics break down because agent-executed tasks look indistinguishable from legitimate developer work. The $100M seed (and In-Q-Tel backing) signals investor conviction that AI-native security is now table-stakes, not an add-on. Enterprises shipping agentic workflows need to reckon with governance layers that understand the semantics of AI decision-making in real time.