Capital signal
Obsidian Raises $85M for AI Agent Security
Obsidian Security raised an $85 million Series D at an $1.1 billion valuation, signaling dedicated capital for securing enterprise AI agents and non-human identities.
Obsidian Security has announced an $85 million Series D financing at an $1.1 billion valuation. The completed round suggests that, as enterprises address access-control issues created by AI agents and non-human identities, investors are beginning to treat this security need as a distinct product and budget category rather than merely an extension of conventional SaaS security.
Funding puts agent security on a distinct capital track
In its August 4 announcement, Obsidian Security confirmed an $85 million Series D at an $1.1 billion valuation. The company describes its business as protecting AI agents and non-human identities in enterprises. For the security market, the round shows that investors are willing to fund vendors focused specifically on the identity, authorization, and access risks surrounding autonomous enterprise actors. Obsidian had previously announced a $90 million Series C in April 2022; the new round moves it into another stage of capital-backed expansion.
The security boundary is moving beyond human accounts
As enterprises deploy AI agents, the actors able to call systems, retrieve data, or act on behalf of employees are no longer limited to human user accounts. Obsidian's framing places AI agents alongside non-human identities, pointing to a governance surface that includes authorization, continuous monitoring, and response for automated actors. The editorial inference is that demand will depend on whether agents receive real cross-system execution rights: the closer those permissions are to production operations, the more identity governance may become an operational control point rather than a compliance add-on.
The financing will test whether buying follows deployment
The round gives Obsidian resources to invest in product development and go-to-market, while raising expectations for commercial conversion. The strongest countercase is that funding and valuation alone do not establish broad enterprise purchasing of agent-security tools; many organizations may initially extend existing identity and SaaS-security platforms instead. Even so, the $1.1 billion valuation shows that investors are already assigning value to earlier governance of agent permission risk.
What to watch next
Observable next evidence includes disclosed large-enterprise customers, the scale of agent-security deployments, retention or revenue metrics, and whether identity-management and security-platform vendors release comparable non-human identity capabilities. Parallel movement across those indicators would strengthen the case that agent security is becoming a separate procurement layer; absorption by incumbent platforms would weaken it.