Okta acquires startup Permiso for $200 million to secure AI agents in the cloud

Okta is buying the startup Permiso for ~$200 million. The deal gives the company identity threat detection technology, primarily for protecting AI agents and non-human accounts (service accounts, API keys) in the cloud. Demand for this kind of protection has surged as enterprises deploy agentic AI systems with broad access privileges.

AI-processed from TechCrunch; edited by Hamidun News
Okta acquires startup Permiso for $200 million to secure AI agents in the cloud
Source: TechCrunch. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceTechCrunch ↗
Editor-in-Chief
Deal valueabout $200 million (according to TechCrunch's source)
Announcement dateJuly 30, 2026
Key asset acquiredidentity threat detection technology

On July 30, 2026, Okta announced the acquisition of Permiso, a startup specializing in identity threat detection in AI environments. According to TechCrunch's source, the deal is worth about $200 million.

What Okta gets

The deal gives Okta identity threat detection technology. Permiso focuses on protecting non-human identities (NHI) — service accounts, API keys, and AI agents that operate in multi-cloud environments without human involvement.

  • Deal value: about $200 million (according to TechCrunch's source)
  • Announcement date: July 30, 2026
  • Key asset acquired: identity threat detection technology
  • Permiso's focus: non-human identities — AI agents, service accounts, and API keys in the cloud
"The deal gives Okta identity threat detection capabilities as enterprises look to secure AI agents and other non-human identities in cloud environments,"

TechCrunch reports.

Multi-cloud environments add a particular complication: a single AI agent can have access to dozens of systems at once through API tokens and service accounts. If such an agent is compromised, the attacker gains privileges comparable to those of a system administrator, and standard SIEM tools may miss the anomaly because they cannot assess the behavior of autonomous agents.

Why AI agents have become the main identity threat

AI agents open a fundamentally new attack vector in enterprise security. Traditional identity and access management (IAM) tools were built for people: they track passwords, MFA, and sessions. Non-human actors — API keys, service accounts, CI/CD tokens — outnumber employees many times over in a large company's infrastructure, and most IAM platforms cannot monitor their behavior in real time.

According to TechCrunch, growing enterprise demand for protecting AI agents and non-human identities in cloud environments was the main rationale for the deal. For a company that already holds a leading position in enterprise IAM, buying Permiso means expanded functionality: a move from managing access to actively detecting threats.

What this means

The acquisition of Permiso for ~$200 million sets a market benchmark: large security vendors are starting to buy up niche AI-security startups without waiting for the segment to mature. For enterprises deploying agentic AI systems, it is a signal that protecting non-human identities is becoming a mandatory element of corporate security rather than an optional add-on.

Frequently asked questions

What are non-human identities (NHI)?

Non-human identities are accounts and tokens that belong not to people but to programs: AI agents, service accounts, API keys, and CI/CD systems. In a large enterprise's infrastructure, they can outnumber employees by 10–50 times, and traditional IAM tools are not designed to monitor their behavior.

Has the deal value been officially confirmed?

The figure of about $200 million comes from TechCrunch's source — Okta had not officially disclosed it at the time of publication.

Story

Key Questions Answered

What are non-human identities (NHI)?

Non-human identities are accounts and tokens that belong not to people but to programs: AI agents, service accounts, API keys, and CI/CD systems. In a large enterprise's infrastructure, they can outnumber employees by 10–50 times, and traditional IAM tools are not designed to monitor their behavior.

Has the deal value been officially confirmed?

The figure of about $200 million comes from TechCrunch's source — Okta had not officially disclosed it at the time of publication.

What Okta gets?

The deal gives Okta identity threat detection technology. Permiso focuses on protecting non-human identities (NHI) — service accounts, API keys, and AI agents that operate in multi-cloud environments without human involvement. Deal value: about $200 million (according to TechCrunch's source) Announcement date: July 30, 2026 Key asset acquired: identity threat detection technology Permiso's focus: non-human identities — AI agents, service accounts, and API keys in the cloud "The deal gives Okta identity threat detection capabilities as enterprises look to secure AI agents an

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Autonomous AI Execution · Nomad Agent

Delegate Deep Research, Coding & Multi-Step Workflows to Nomad Agent

Deploy autonomous task execution with Nomad Agent (agent.nomadnet.ai) — automated web research, data synthesis, code generation, and MCP tool orchestration.

Launch Nomad Agent →Nomad AI Workspace
What do you think?
Loading comments…