As generative AI tools spread across HR, recruiting, scheduling, and internal operations, a new study from HERE Enterprise warns that many organizations are paying a "toggle tax": the time employees spend switching between AI systems, verifying outputs, correcting errors, and translating between tools can rival the time saved by the technology itself.

The finding challenges the common assumption that AI automatically lifts productivity. According to the research, users report spending as much effort managing workplace AI as they gain from it. Some of that burden is visible—logging into multiple platforms, copying data between a chatbot and a spreadsheet—but much of it is cognitive: deciding whether to trust a draft, double-checking facts, rewriting tone, or reverting to manual work when the AI misunderstands context.

What the toggle tax looks like in daily work

In practice, the toggle tax appears in small, repeated frictions. A recruiter may use one tool to summarize résumés, another to draft outreach, and a third to log activity in an applicant tracking system. Each handoff requires review and re-entry. A people manager may ask an assistant to produce a performance review draft, then spend nearly as long editing it as writing from scratch. HR teams experimenting with AI for policy questions face a similar pattern: the system produces a plausible answer, but the professional must still verify it against current regulations and internal rules.

The research suggests that these costs are often underestimated because they are distributed across the workday rather than appearing as a single line item. Leaders see the promise of automation but not the accumulated minutes spent switching context, correcting outputs, or explaining tasks repeatedly to tools that lack organizational memory.

Why adoption outpaces readiness

Part of the problem is that workplace AI has been adopted faster than integration and training. Many organizations license multiple point solutions that do not share data, forcing employees to act as the connective layer. The HERE Enterprise study points to a mismatch between how vendors demonstrate the technology—usually on a clean, single-task basis—and how real work happens across messy, interconnected systems.

Another factor is trust calibration. Employees quickly learn that AI is confident but sometimes wrong, especially in HR-adjacent tasks involving nuance, policy exceptions, or sensitive employee situations. The safe response is to verify everything, which consumes the time the tool was meant to save. When verification is required, the net benefit can shrink dramatically.

This does not mean workplace AI is useless. The study indicates that the problem is less about the underlying models and more about how organizations deploy them: without clear workflows, integration, and guidance on when to trust or override outputs, employees are left to manage complexity on their own.

Turning a tax into a return

Organizations that reduce the toggle tax tend to focus on a few practical steps. First, they consolidate tools around specific workflows instead of allowing every team to pilot a different assistant. Fewer systems mean fewer handoffs. Second, they invest in integration so that AI tools can read and write directly into the systems where work already happens, such as the HRIS, ATS, or internal knowledge base. Third, they create clear policies for verification: not every draft needs a full audit, but high-risk outputs—like compensation guidance or termination language—always do.

Training also matters. The most effective programs teach employees not just how to prompt, but how to evaluate responses, recognize typical failure modes, and recover quickly. This kind of literacy reduces the mental overhead of using AI and helps teams separate tasks that genuinely benefit from assistance from those that are still faster to do manually.

For HR leaders, the research is a reminder that adopting AI is not the same as capturing its value. The metric to watch is not how many employees have access to AI, but whether the total time and quality of their work actually improves after the tool is introduced. A candid audit of where AI saves time and where it adds hidden steps can reveal whether the organization is earning a return or simply paying a toggle tax.

As remote and hybrid teams increasingly rely on digital tools to stay connected, managing that hidden overhead becomes even more important. A well-integrated workplace AI strategy—like a well-designed remote-work platform—should reduce friction rather than add to it. That is the standard platforms such as XMF encourage for the teams they support.

Originally published by XMF, inspired by publicly reported industry news.