It's 11pm on a Tuesday. A mid-level manager is lying awake, replaying a tense conversation with her director and dreading tomorrow's performance review.
But she isn’t going to consult her organization’s EAP. She doesn't even know where to find it. And she's not going to open the wellness app her company rolled out last spring and hasn’t promoted since.
Instead, she tells her fears to ChatGPT, and spends the next hour working through her anxiety with a general-purpose AI that was built to write cover letters and summarize PDFs.
This isn't a hypothetical. It's happening all across workforces right now—quietly, frequently, and almost entirely outside the organizational view.
It’s not necessarily detrimental to employee wellbeing, but it pulls workforces away from the plethora of resources, point solutions, and support that are in place to help them. And while organizations continue to debate whether AI belongs in their wellness programs, their employees have already answered that question.
Nearly half of U.S. adults used a general-purpose AI tool for psychological support in the last year alone. Yet only a fraction used a tool built intentionally for mental health, and that number drops even further when looking specifically at workplace mental health benefits.
The use of AI chatbots isn't a fringe behavior confined to tech-savvy early adopters. It's swiftly becoming a widespread, accelerating pattern, and it’s spreading across every type of workforce—from hourly workers and remote employees to frontline staff and senior leaders alike.
In other words, the gap between what employees are doing and what their organization acknowledges is enormous, and it's widening every day.
To effectively bridge the gap, it helps to first understand why general-purpose AI became the default mental health resource for so many without any promotion, onboarding, or communication.
It's always available. There are no redirects, broken links, or forgotten logins. The moment an employee needs support, it's there. Most traditional mental health benefits can't say the same.
It feels low-stakes. Talking to a chatbot doesn't feel like "using your EAP." For many employees, contacting a counselor through an employer-sponsored program raises uncomfortable questions about confidentiality and workplace stigma. General-purpose AI carries none of that perceived risk.
It bypasses stigma entirely. When something doesn't feel like "getting help," people will wholeheartedly jump in. That frictionless experience sidesteps the stigma barrier that causes so many people to never engage with traditional resources.
What this reveals is a fundamental access and awareness problem with traditional wellness benefits. It's not that employees don't need support, but rather the support organizations offer hasn't been meeting employees where they are.
When organizations learn that employees are using general AI tools for mental health support, the instinctive response is often to restrict it.
This response is understandable, but it doesn't work—and may make things worse.
Employees who are turning to ChatGPT at 11pm for support aren't doing it because they’re told to, but because it's the most accessible option they have in that moment. A policy restriction doesn't change that equation; it just ensures that if an employee is struggling, they're more likely to do so in ways the organization can't see, support, or respond to.
Organizations that respond to this behavior with "we don't allow that" are opting out of the conversation at exactly the moment when their employees need them most. The question isn't whether employees will seek AI-assisted support; it’s whether the organization will play any role in making that experience safer and more convenient.
If you're still framing this as a debate about whether AI belongs in your wellness program, it's time to change the question.
The more productive ask is: how do you make AI work responsibly for your employees and your organization?
🔎 Start by meeting employees where they are. Rather than treat AI-assisted mental health support as a threat, acknowledge it as part of the modern mental health landscape. In response, tailor your resources and support guides around what employees are actually doing, not just what you’ve historically offered.
🔀 Assess whether your platform can integrate purpose-built tools. There's a meaningful difference between an AI chatbot that happens to be used for emotional support and a wellness platform that prioritizes mental health and is built with clinical oversight, privacy protections, and evidence-based frameworks. Ensuring your wellness platform can seamlessly house these tools without creating friction gives employees the accessibility they're seeking, without the risks.
👩⚕️ Pair AI accessibility with human escalation pathways. The most effective wellness programs treat AI health support as a point of entry, not a destination. When an employee consults an AI guide within their wellness platform and it signals they may need more assistance, there should be a clear, low-friction path to support.
💌 Communicate proactively and specifically. Most employees who turn to ChatGPT for support aren't aware that a safer alternative exists in their wellness program. Don't assume they'll find it on their own; instead, outfit your wellness platform so that it communicates a simple message: here's a tool that does what you're already looking for, with the protections you deserve. Delivered clearly and repeatedly, that message can shift behavior.
📱 Make benefits as easy to access as a chat window. The EAP that requires three phone calls and a two-week wait isn't even close to competing with ChatGPT. If your wellness benefits aren't immediately accessible and available on employees' own terms, friction is doing the work of disengagement for you.
💡 Turn your employees' AI fluency into a benefits engagement advantage. Your employees are already comfortable talking to AI, and that's actually an asset you can use. By deploying an AI-powered tool that helps employees navigate your wellness platform, answer questions about available resources, and find the right program for what they're going through, you stop fighting the instinct to "just ask the AI" and start channeling it.
Not all AI-assisted wellness program tools are created equal. When evaluating platforms, look for one that has:
The debate inside many HR teams is framed as: Should we allow AI in our wellness program?.
The question that actually matters is: Do we want our employees using ungoverned AI tools in moments of genuine vulnerability, or do we want to offer them something better?
The window to get ahead of this is narrowing. The longer organizations wait, the more entrenched the behavior becomes, and the harder it is to redirect employees toward safer, clinically informed alternatives within their wellness programs.
Most workforces aren't at the beginning of this shift. They're in the middle of it.
WellRight's approach to mental wellbeing is built for exactly this moment, combining accessibility with clinical integrity and designed to meet employees where they are while guiding them toward the support they actually need.
Ready to see what that looks like for your organization? Request a demo today.