24% of Departed Employees Retain Access to Company AI Tools

Nearly a quarter of former employees — 24% — said they could still use a company AI tool or assistant after leaving their job, according to a survey of 1,000 U.S. adults published in October by managed IT and cybersecurity provider Sagiss. Another 16% reported that company information remained in an AI tool or chat history they had used for work.

The survey, conducted in September through Pollfish, covered adults who had left jobs involving company technology within the previous five years. The findings are self-reported experiences, not verified audits. For HR leaders, they point to a concrete process failure: offboarding checklists that disable email and standard accounts while leaving AI-era access — and the data inside it — unaddressed.

What does 'access' actually mean here?

The survey did not establish what kinds of AI accounts former employees could still use, according to Travis Springer, president of Sagiss. Some may have been company-managed accounts that were never disabled; others may have been personal AI accounts used for work. The survey could not determine which situation is more common, and the distinction matters, because the two problems require different responses.

A company can disable an account it manages. Information already copied into a personal AI tool may remain available through an account the employer does not control.

"Disabling an account is only part of the job. You also have to think about what company information has already been put into these tools and who controls it," Springer said.

The exposure extends past standalone chatbots. Security consultant Dave Shackleford identified additional risks in AI-enabled SaaS applications (software delivered over the cloud rather than installed locally), which can query business systems, modify records and invoke APIs on a user's behalf. He also flagged service accounts, API tokens and OAuth grants (mechanisms that let one application act on data in another) as blind spots. Some are long-lived or poorly tracked, and AI agents can inherit departing employees' permissions — widening the access available through those connections.

The implication for HR and IT alike: inventory embedded AI capabilities, connected agents and nonhuman identities, not merely the applications on the official approved list.

Who is responsible for cutting off access?

The problem is often organizational rather than technical. Springer said Sagiss regularly discovers accounts belonging to departed employees during routine client reviews because IT was never told those employees had left.

"We regularly review clients' inactive accounts and identify people who left the company without IT ever being notified," he said.

He described a handoff in which each department holds one piece of the puzzle: "HR knows the employee is leaving, the manager knows which tools that person used, and IT manages the accounts it knows about. Sometimes those pieces never come together."

The numbers back him up. Among respondents who had used seven or more work accounts, 87% said at least one still worked after departure. Even among those with only one to three accounts, 71% reported the same problem. And 33% of all respondents said responsibility for terminating their access was unclear or appeared to be nobody's job.

How are employers finding 'shadow AI'?

Some CIOs are already mapping informal AI use. At Flexera, CIO and CISO Conal Gallagher said the company surveyed employees directly to surface AI tools adopted without formal IT approval and identify which ones genuinely helped people work.

"Our first step was to bring informal AI use cases into the open by engaging employees," Gallagher said.

MoonPay, a cryptocurrency payments company, took a more technical route. CIO and CISO Doug Innocenti said the firm used AI security software to examine applications and their underlying models, including how providers might handle company data. "We were able to perform an independent review on tools and understand which ones we wanted to pick," he said.

The methods differ. Both address the same prerequisite for reliable offboarding: knowing which applications employees use and where company information flows.

Why single sign-on is not a finish line

Single sign-on (one central login governing many applications) simplifies offboarding, but Microsoft notes that shutting off an employee's access does not necessarily end every active session immediately. Some applications still accept previously issued tokens or maintain their own sessions.

Springer recommends checking applications outside SSO, shared credentials and lingering sessions, plus reviewing what information employees stored in AI services and whether administrators retain control over it. He advises documenting which accounts were disabled, when, and who verified the changes — and investigating delays, missed applications and unexpected post-departure logins. In his view, HR should see the offboarding process through to the end, even where IT and security do the technical work.

"You should be able to confirm someone is off the systems instead of finding out weeks later that an account still works," Springer said.

As AI agents and embedded copilots multiply across the average software stack, HR teams that treat access termination as a documented, cross-functional handoff — rather than a closed ticket — will be the ones who can answer that question with confidence.