Okta Beats Earnings Forecasts as Companies Rush to Secure AI Agents
The identity security firm raised its annual outlook after a stronger-than-expected quarter, with demand rising from businesses trying to control who, and what, can access their systems.
Key points
- Okta reported quarterly financial results that beat Wall Street's expectations, sending its share price sharply higher.
- The company raised its full-year revenue outlook, signalling confidence in sustained demand.
- A growing driver of that demand is the rise of AI agents, software programs that act autonomously on behalf of a business, which need the same kind of identity controls as human employees.
- Securing these so-called "non-human identities" is becoming one of the fastest-growing problems in corporate security.
Okta sells software that controls who is allowed into a company's digital systems. Think of it as the bouncer checking IDs at the door of every app a business runs. This past quarter, more companies than expected came knocking for that service, and Okta's results reflected it, as first reported by SecurityWeek.
Shares jumped after the company disclosed it had beaten its own financial targets and, more significantly, raised its outlook for the rest of the year. Investors read that as a sign the demand driving those results is not a one-off.
Why are businesses suddenly buying more of this?
Two forces are pressing at once. First, remote work and cloud computing have meant that employees, contractors, and partners all log into company systems from everywhere, making it harder to know whether any given login is legitimate. Second, and newer, companies are deploying AI agents.
An AI agent is a piece of software that can book meetings, send emails, query databases, or run reports without a human clicking anything. It acts. That means it needs access credentials, essentially a digital pass, just like a person does. Right now, most organisations have very little visibility into how many of these automated programs are running inside their systems, what they can access, or what would happen if one were taken over by an attacker.
Security researchers call these "non-human identities," and the number of them inside a typical large company now dwarfs the number of human employees with login credentials. Controlling them is, bluntly, an unsolved problem for most IT teams.
Should ordinary customers or employees be worried?
Not directly from this news story, which is about a company's finances rather than a breach. But the underlying issue matters to anyone whose data sits inside a business using AI tools.
If an AI agent has broad access to customer records and that agent's credentials are stolen or misused, the effect on real people can look identical to a traditional data breach. Names, addresses, financial details, medical records: whatever the agent could reach, an attacker could reach too.
The practical takeaway for employees: if your workplace is rolling out AI assistants or automation tools, it is reasonable to ask what data those tools can see and who is watching what they do. That is not paranoia; it is the same question your IT team should already be asking.
For Okta itself, the numbers suggest the market agrees the problem is real and growing. Whether the company can stay ahead of that growth, particularly as rivals sharpen their own identity offerings, is the quieter question behind the strong quarter.
