Neo launches with $100M for AI-agent security
Neo has emerged from stealth with $100 million to help enterprises identify and control the permissions of AI agents and AI-enabled applications.
Enterprise software is gaining autonomous capabilities faster than many security teams can understand what those applications are allowed to do.
What happened
Neo emerged from stealth with $100 million across seed and Series A financings from Andreessen Horowitz, Bessemer Venture Partners, Craft Ventures and Merlin Ventures.
Founded by former SentinelOne executives, the company is building a security platform for AI agents and conventional applications that gain agent-like functionality. Neo aims to identify which systems are operating, what data they can access, which actions they can perform and whether an activity was initiated by a person, application or model.
Why it matters
An enterprise may approve a familiar software product and later discover that a routine update has added model access, automation or new permissions. Traditional application inventories do not necessarily capture how those capabilities change operational risk.
The problem becomes more serious when agents can read sensitive information, modify records or trigger external actions. Security teams need visibility into both identity and behaviour, along with policies that can stop an agent from exceeding its intended role.
The bigger picture
AI security is moving beyond protecting models from attacks. A large new market is forming around controlling the software actors built with those models.
Neo’s experienced founders and unusually large launch financing signal investor confidence, but the competitive field is becoming crowded. Identity vendors, cloud platforms and existing cybersecurity companies are all extending products into agent governance. Neo must prove that its approach can become a distinct control layer rather than a feature absorbed by larger platforms.