Okta announced Thursday that it has agreed to acquire Permiso Security, a startup focused on AI identity security, in a deal valued at just under $200 million. The acquisition underscores Okta's push to expand beyond traditional identity and access management into the rapidly evolving world of AI agents and machine identities. As enterprises increasingly embed autonomous software into their operations, the need to secure non-human identities has become a critical priority.

Deal Terms and Expected Closing

While Okta did not disclose the financial terms of the transaction, a source familiar with the matter said the acquisition is valued at just under $200 million and is structured as an almost all-cash deal. When asked directly about the figure, Okta's CEO Todd McKinnon did not dispute it, though the company declined to comment on specific deal terms. The transaction is expected to close in the third quarter of Okta's fiscal 2027, subject to customary closing conditions.

The price tag represents a significant return for Permiso's early investors. The Palo Alto-based startup had raised approximately $29 million to date, including an $18.5 million Series A round in April 2024 led by Altimeter Capital. People familiar with that financing said the round valued Permiso at about $80 million on a post-money basis. With this acquisition, Permiso's backers are seeing a valuation more than double the company's last private market assessment.

What Permiso Does

Permiso emerged from stealth in 2022 with a focus on helping security teams spot suspicious activity in cloud environments after users or applications have already been granted access. In the cybersecurity industry, this is often referred to as identity threat detection and response (ITDR). Traditional identity tools verify credentials at login, but they often fail to monitor what happens afterward. Permiso's software continuously analyzes behavior, looking for anomalies that might indicate a compromised identity or a malicious insider.

More recently, Permiso has expanded its platform to monitor AI agents and other machine identities. AI agents are software programs that perform tasks autonomously, often interacting with cloud services, APIs, and data stores. As these agents proliferate, they create new attack surfaces. Permiso's technology helps organizations detect when an AI agent is behaving unexpectedly or when a machine identity has been stolen and used for unauthorized access.

In April 2025, Permiso introduced SandyClaw, a platform designed to analyze AI agent skills in a sandboxed environment. The goal, according to the company, is to identify malicious behavior before agents are deployed in production. By simulating an agent's actions and inspecting its skills, SandyClaw can uncover attempts to exfiltrate data, escalate privileges, or manipulate other systems.

Founders and Background

Permiso was co-founded by Paul Nguyen and Jason Martin, both former FireEye executives. FireEye is a well-known cybersecurity firm that specializes in advanced threat detection and incident response. Nguyen and Martin brought deep expertise in cloud security and threat intelligence to their new venture.

Nguyen, who serves as CEO, previously led security research and product development at FireEye. Martin, the CTO, had a similar background in building detection engines and scalable security platforms. Their experience at FireEye gave them a front-row seat to the growing challenge of identity-based attacks. Attackers increasingly bypass traditional defenses by stealing valid credentials and moving laterally through cloud infrastructure. Permiso was designed to address that specific gap.

According to the startup's public materials, its platform is used by security operations teams in a range of industries, including technology, finance, and healthcare. The company has also built a threat research arm that publishes analyses of emerging attack techniques related to cloud identities and AI agents.

Strategic Rationale

Okta's acquisition of Permiso reflects a broader shift in the identity management market. For years, the industry focused on verifying users at the point of login. But the rise of cloud-native applications, DevOps pipelines, and AI-driven automation has changed the calculus. Enterprises now need to continuously monitor what users, applications, and AI agents do once they have gained authorized access to a network environment.

This shift has intensified competition to secure machine identities. Unlike human users, machine identities do not have the ability to change passwords or respond to multi-factor authentication prompts. They often operate behind the scenes, making them an attractive target for attackers. Industry analysts estimate that the number of machine identities in enterprise environments can outnumber human identities by 10 to 1 or more. Managing and securing those identities has become a major pain point for organizations.

By adding Permiso's ITDR capabilities, Okta can offer customers a more comprehensive identity security fabric. Rather than only provisioning and authenticating identities, Okta can actively detect and respond to threats that arise after authentication. The company's chief product officer, Ely Kahn, said in a prepared statement: "Permiso will extend Okta’s identity security fabric with proven identity threat detection and response capabilities, and an incredible threat research and security team that will advance Okta’s threat detection and prevention capabilities."

Market Context and Competition

The deal comes at a time when the cybersecurity market is consolidating around identity. Okta already competes with Microsoft, Ping Identity, and others in the identity and access management space. But the acquisition of Permiso signals a more aggressive push into the ITDR segment, which has traditionally been occupied by smaller, specialist vendors.

Microsoft has invested heavily in its own identity threat detection capabilities, such as Microsoft Defender for Identity. Other players, including CrowdStrike and Palo Alto Networks, have also expanded into identity threat detection as part of their broader security platforms. Okta's move suggests that it intends to compete head-on in this emerging category.

The focus on AI agents is particularly timely. In just the past year, enterprises have moved from experimenting with large language models to deploying AI agents that can take meaningful actions autonomously. These agents interact with business applications, databases, and customer systems. If an attacker compromises an AI agent, they could potentially cause significant damage without needing direct access to a human user's account. Permiso's SandyClaw addresses this risk by evaluating agent skills before deployment and monitoring them in runtime.

Financial Details and Funding History

Permiso's investors include Altimeter Capital, a well-known venture capital firm focused on growth-stage technology companies. The funding history reflects a typical venture trajectory: an initial seed round to get the company off the ground, followed by a Series A to scale product development and go-to-market efforts. With a total raise of around $29 million, Permiso was relatively capital-efficient compared to many cybersecurity startups.

The reported acquisition price of just under $200 million would represent a significant multiple on total funding. It also highlights the premium acquirers are willing to pay for technology that bridges the gap between identity security and AI security. In a market where trust in AI is a growing concern, having visibility into what AI agents are actually doing is considered a critical capability.

What Customers Can Expect

Okta has indicated that Permiso's technology will be integrated into its existing identity security offerings. Customers can expect to see new features in the Okta platform that provide continuous threat detection across human and machine identities. For large organizations using Okta's workforce identity solutions, this could mean better visibility into suspicious activity across their entire cloud environment.

The acquisition also brings over a team of threat researchers and security engineers who will join Okta's security organization. This team will likely contribute to Okta's broader threat intelligence and incident response efforts. For current Permiso customers, Okta has not yet announced specific product roadmap changes, but it has committed to maintaining and supporting the Permiso platform.

The deal is expected to close in the third quarter of Okta's fiscal 2027, which runs from August to October 2026. Until then, Permiso will continue to operate as an independent entity, and its product will remain available to existing customers. The companies have not said whether Permiso will be rebranded or folded into Okta's existing product line immediately after the close.

For the broader cybersecurity industry, this acquisition serves as a signal that AI security is no longer a niche concern. It is becoming a core part of identity management, and large vendors are willing to pay significant sums to acquire specialized capabilities. As enterprises continue to deploy AI agents across their operations, the ability to secure those agents will likely become a key differentiator for identity platforms.

Source:TechCrunch News