Okta’s selloff looks wrong because the market is still treating this as a mature identity vendor just as the company is turning AI-agent security into a real product category. The key point is not hype around AI in the abstract; it is that Okta has already launched Okta for AI Agents – Core for regulated environments and is shipping agent-to-agent controls, token exchange tooling, and developer features through Auth0. That matters because identity is one of the few layers every autonomous software agent will need, especially in FedRAMP and HIPAA settings where governance is not optional. At $138.63, the stock is expensive on backward-looking multiples, but the tape is missing that a new identity land grab is starting in a market Okta already knows how to sell into.
The business underneath the AI narrative is stronger than the “slow grower” label suggests. Revenue is still growing 11.8% year over year on a $2.92 billion base, and earnings are inflecting much faster, with EPS growth of 688.2% and net income growth of 739.3%. That is exactly the kind of setup that can support a premium multiple when a company is moving from scale-at-all-costs to profitable expansion. Public filings also showed GAAP operating income of $56 million in the latest quarter, or 7% of revenue, up from 6% a year earlier, so this is not a story stock detached from operating progress.
The product side is where the market looks too dismissive. Okta launched Okta for AI Agents – Core on June 25 for regulated environments, then followed with agent-to-agent connection controls that let admins define invocation rights, scope, and session duration. Auth0 has also positioned autonomous agents as first-class identities, which gives Okta a developer channel into agentic applications instead of relying only on enterprise CIO budgets. If AI agents become a serious enterprise deployment wave, identity governance is not a nice-to-have feature bolted on later; it is core infrastructure, and Okta is already productizing it.
The stock action before today’s drop also argues this is a reset, not a breakdown. OKTA is still up 65.7% year to date, outperforming the Technology sector by 37 percentage points, and it remains above its 20-day, 50-day, and 200-day moving averages. The TickerSpark Score sits at 79 overall, with Growth at 90, Financial Health at 84, and Momentum at 100. That profile does not read like a broken software name. It reads like a company with strong execution getting hit by a sentiment wobble even as recent news flow stayed strongly positive.
The cleanest pushback is valuation. A 99.02x trailing P/E and 63.00x EV/EBITDA leave very little room for disappointment, especially when revenue growth is 11.8% rather than the 30%-plus pace investors usually associate with AI winners. Compared with FFIV, which trades at 25.47x earnings with a 22.0% net margin, Okta is clearly the pricier stock for a business still proving out its next leg.
There is also a fair skepticism around whether AI-agent identity is a real revenue driver yet or still mostly a well-timed product narrative. Recent insider activity will not calm that concern either: the last roughly 10 reported insider transactions were all sells, totaling 73,499 shares worth $10.64 million. Even so, that counterargument falls short because the market is not being asked to underwrite a pre-revenue moonshot. Okta has beaten EPS estimates in 7 straight quarters, carries a Buy consensus with 37 buys against 12 holds and 2 sells, and is layering the AI-agent opportunity onto an already profitable core business.
That leaves us on the side of buying the dislocation, not the panic. We would rather own OKTA than FFIV right now because Okta offers a live category-creation angle on top of improving profitability, while FFIV looks more like a steadier but lower-upside incumbent. The market is punishing OKTA as if the AI-agent push is marketing fluff, and that looks too cynical given the actual launch cadence and regulated-environment focus.
What we would watch from here is simple: the stock needs to keep holding above its major trend lines, and the next earnings cycle needs to show the same pattern of disciplined execution that produced a 7-for-7 earnings beat streak. If operating leverage stalls or management cannot translate AI-agent launches into bookings language, the premium multiple gets harder to defend fast. Until that happens, this selloff looks more like a chance to accumulate a category leader than a signal to run.
