1Password Buys Apono to Expand AI Access Governance

1Password purchased a startup founded by an Israeli government and military cyber leader to bring credential security, access governance and just-in-time authorization together.

See Also: Privilege Blind Spots: Part 1, Uncover Risk from Siloed Identity Tools

The Toronto-based identity security vendor said buying New York-based Apono will allow 1Password to converge identity, credential management and access governance into a unified framework capable of supporting humans, machines and agents simultaneously, according to CEO David Faugno. Standing privileges and static permissions fall short against AI-powered agents and automated workflows.

“The systems that people are identifying to support this future don’t really exist, and we’re among the many companies in the world that are deploying AI internally and seeing these challenges ourselves,” he told ISMG. “We actually were starting to look at, ‘How do we issue just-in-time access to machines and agents and people that are delegating access to agents in a way that keeps us safe internally?'”

Apono, founded in 2021, employs 80 people and has raised $54.5 million, having last completed a $34 million Series B funding round in November 2025 led by U.S. Venture Partners. The company has been led since inception by Rom Carmel, who spent nearly three years doing security software in the Israel Defense Forces and three-and-a-half years in the cyber division of the Israeli Prime Minister’s office (see: 1Password Acquires Trelica to Boost SaaS Access Management).

How Access Governance, Credential Management Come Together

The company issues access only when needed, Faugno said, with permissions tied to specific tasks, limited in duration and governed by policies that evaluate the intended purpose of the request. Faugno described this as intent-based access control, where access decisions are driven not simply by who or what is requesting access, but by what the requester is attempting to accomplish.

“What Apono brought to the table was really zero standing privilege by default,” Faugno said. “It’s core to their architecture, an access that is not role-based, but is task-scoped, time-bound, and really tied to agent intent. This intent-based policy, with just-in-time delivery and zero standing privilege, was how we thought about the missing piece.”

Organizations increasingly have credentials scattered across code repositories, cloud environments and developer tools, with developers frequently storing static credentials where attackers can easily find them. One of the central goals of 1Password’s unified access strategy is to discover exposed credentials, bring them under governance and replace static access with dynamic, policy-driven authorization.

“Converging into a single control plane and extending your trust from just securing an identity credential to governing what that identity can do for how long, when, and why, that’s really how the things come together,” Faugno said.

Combining the ability to securely store credentials and issue them as needed with Apono’s access governance and just-in-time authorization capabilities allows 1Password to create a unified control plane for access management, Faugno said. With this model, companies can not only secure credentials but also govern what identities are allowed to do, when they can do it and under what conditions.

“A system that people won’t work around, that actually can broker access to the agents, that can really unlock the value that this agentic workflows represent, solves this elusive problem,” Faugno said. “We’re really set up now with these capabilities.”

How Security Can Ensure Agents Remain Within Boundaries

Security leaders want assurance that agents are operating within authorized boundaries, using only approved permissions and performing intended tasks. Without mechanisms for auditing, revoking and controlling agent access, many organizations remain hesitant to fully embrace agentic workflows. Solving this challenge will be essential to unlocking the full value of enterprise AI investments, he said.

“The really powerful agentic deployments are getting stopped because there is no easy way to make sure that agents are acting on behalf of users in a predictable way with the access that they’re given,” Faugno said. “So you’re seeing a lot of security teams jump in front of agent deployments in businesses.”

Organizations should evaluate whether a requested action aligns with the agent’s intended purpose, which Faugno said can address concerns such as prompt injection attacks, agent drift and unintended behavior. By continuously validating intent and limiting permissions to specific tasks, organizations can reduce the likelihood that compromised or malfunctioning agents will gain excessive access, he said.

“Moving away from static permissions to dynamic issuance just in time is really a way to test and assert that this access makes sense, and it makes sense still, and this is how long it makes sense for, and this is how we’re going to make sure we manage that,” Faugno said.

Combining Apono’s access management technology and 1Password’s credential security capabilities positions the company to play a much larger role in identity and access management, Faugno said. Rather than serving as a stand-alone password manager, Faugno said 1Password aims to become a core platform for managing access across humans, machines and AI agents.

“1Password now has moved from really acting on the periphery to being a core system of access and access governance for our customers,” Faugno said. “That allows us to really play a much different game going forward.”