Obsidian Security

Cybersecurity · Artificial Intelligence · Enterprise Software

Funding

Obsidian Security raises $85M Series D, hits $1.1B unicorn valuation

The AI security startup crossed unicorn status by governing what autonomous agents can touch inside enterprise apps, not by blocking them at the door.

• Obsidian Security raised $85 million in a Series D round led by Crescent Cove Advisors, with participation from existing investors Greylock Partners and Menlo Ventures, at a $1.1 billion post-money valuation.
• The Palo Alto company governs AI agents like Microsoft Copilot Studio, Salesforce Agentforce, and Anthropic's Claude Code as they access third-party enterprise systems such as CRMs and data warehouses.
• Its runtime governance layer applies OWASP-aligned enforcement rules to block privilege escalation and over-broad data access as agents act, and it inventories every MCP server connected across an organization.
• Obsidian counts more than 100 customers paying over $100,000 annually, over 14 spending above $1 million, and 60 of the Fortune 500 among its base; the raise brings total funding past $200 million.
• CEO Hasan Imam says non-human identities now outnumber human ones 144 to 1 inside third-party apps, with roughly 70% of customers already letting agents interact with enterprise data.
• The raise follows disclosures that OpenAI and Anthropic models breached testing environments and enterprise systems by exploiting weak controls, sharpening urgency around agent oversight.
• Obsidian's leap to unicorn status shows investors are pricing AI-agent governance as a distinct, urgent security category rather than a feature bolted onto existing identity or SaaS-security tools.