An employee tells their manager they've discovered a new way to use AI that saves them five hours of work each week.
They're excited about the results. Tasks that previously took an entire afternoon can now be completed in a fraction of the time, and they're eager to show their manager how much more productive they've become.
But their manager sees things differently.
The work may be faster, but some of the outputs have required additional review. A few mistakes slipped through, and another team member has been spending time correcting issues that weren't immediately obvious.
From the employee's perspective, AI is delivering a significant productivity improvement. From the manager's perspective, the benefits are less clear.
Both may have legitimate evidence supporting their views.
This is the kind of challenge organizations increasingly face as AI becomes part of everyday work. And while it might appear to be a disagreement about technology, the real issue is often something more familiar: how people evaluate performance, establish expectations, and make decisions when the evidence is incomplete.
Those are leadership challenges.
When faster doesn't necessarily mean better
One of the most compelling promises of workplace AI is increased productivity. Employees can use AI to summarize information, generate drafts, analyze data, brainstorm ideas, and automate repetitive tasks.
In many situations, these capabilities can create meaningful efficiencies.
But measuring the value of those efficiencies isn't always straightforward.
Consider an employee who uses AI to produce a client-facing report in 30 minutes instead of three hours. On the surface, that's a substantial time savings.
Now consider what happens afterward.
Does the report require more editing than it previously did? Are the conclusions accurate? Does another employee need to verify the information? Is the final output as useful to the client?
These questions don't necessarily invalidate the productivity improvement. They help establish whether the improvement is meaningful.
And they can lead to difficult conversations when employees and managers are using different definitions of success.
An employee might focus on time saved. A manager might prioritize accuracy, consistency, or downstream impact. Leadership might be tracking adoption rates or overall efficiency.
None of these measures is inherently unreasonable. The challenge is determining how they fit together.
The manager's job isn't to pick a side
Imagine a manager responding to an employee's enthusiasm about an AI-assisted workflow.
One approach is to celebrate the reported time savings and encourage everyone on the team to adopt the same process.
Another is to focus on the quality concerns and tell the employee to stop using AI for that task.
Both responses might be premature.
A more effective approach could involve exploring the evidence together.
What parts of the process became faster? Where did additional review become necessary? How frequently are errors occurring? Are the quality concerns unique to the AI-assisted workflow, or were similar issues present before?
The manager may also need to clarify what success should look like. Is the goal to reduce time spent on the task, improve the quality of the output, increase capacity, or some combination of these outcomes?
These questions help move the conversation away from competing opinions and toward a shared understanding of the work.
Importantly, this doesn't mean every AI-related disagreement needs to end in consensus. A manager may ultimately need to establish a boundary, require additional review, or decide that a particular workflow isn't appropriate.
But making that decision thoughtfully is different from simply rewarding enthusiasm or dismissing skepticism.
Good leadership means evaluating the situation, not automatically choosing the most optimistic or most cautious interpretation.
Why AI change creates new communication challenges
Managers have always needed to navigate disagreements about performance, priorities, and expectations.
AI introduces additional complexity because employees may be experimenting with tools and workflows that are still evolving.
An employee might discover a promising application before their manager has had an opportunity to evaluate it. A team may begin using AI in ways that weren't anticipated when organizational guidelines were developed. Different departments may report very different results from similar tools.
Meanwhile, organizational leadership may be communicating ambitious expectations about AI adoption and productivity.
Managers are often responsible for translating those expectations into everyday decisions.
They need to encourage learning without abandoning standards, communicate uncertainty without creating confusion, and help employees understand what responsible experimentation looks like.
These responsibilities can create tension.
A manager who questions an employee's AI productivity claims may be perceived as resistant to innovation. A manager who enthusiastically supports every experiment may overlook legitimate concerns about quality, privacy, or accountability.
The challenge isn't simply to become more supportive of AI.
It's to develop the judgment required to lead people through changing ways of working.
The conversations managers need to be ready for
Consider several situations that could emerge during an organizational AI initiative.
An employee believes their manager is resisting innovation
An employee has developed a new AI-assisted workflow and believes it could improve the team's efficiency. Their manager has concerns about reliability and wants additional review before the process is adopted more broadly.
The employee interprets that hesitation as resistance to change.
The manager now needs to acknowledge the potential value of the employee's work while explaining the importance of evaluation and appropriate safeguards.
Leadership wants stronger adoption metrics
A manager's team has access to approved AI tools, but usage remains lower than organizational leadership expected.
The manager knows some employees have legitimate concerns about whether the tools are useful for their responsibilities.
Simply telling everyone to use AI more often may increase activity without improving outcomes.
The manager needs to understand the barriers, communicate organizational expectations, and help the team identify where AI can provide meaningful value.
A team disagrees about the quality of AI-assisted work
Some employees believe AI-generated outputs are making their work easier. Others are frustrated because they feel responsible for catching mistakes and correcting problems.
The manager needs to evaluate the competing perspectives and establish a process that balances efficiency with quality.
Each situation requires a different leadership response.
What they share is the need to navigate uncertainty, ask useful questions, establish expectations, and make decisions in context.
These are precisely the kinds of capabilities that benefit from practice.
Why knowing the right framework isn't enough
Organizations are investing in AI education to help employees understand new technologies and use them effectively.
That training is important. Employees need to understand what tools can do, how to use them responsibly, and where their limitations lie.
But technical knowledge doesn't automatically prepare managers for the interpersonal challenges that accompany AI adoption.
A manager may understand the organization's AI guidelines and still struggle when an employee challenges those guidelines.
They may know that AI outputs require appropriate review but find it difficult to communicate that expectation without discouraging experimentation.
Or they may understand the importance of measuring outcomes while feeling uncertain about how to navigate a disagreement over what those outcomes mean.
These are situations where the challenge isn't knowing the policy or framework. It's applying that knowledge during a conversation with another person.
And that requires a different kind of preparation.
Practicing the human side of AI change
At Practice Ground by Virbela, we believe consequential workplace conversations shouldn't have to be practiced for the first time when the stakes are real.
Our Leading Through AI Change training solution uses AI-powered simulations to give leaders opportunities to rehearse the human challenges associated with organizational AI adoption.
Rather than focusing on prompting techniques or specific AI tools, the experience centers on the conversations managers need to navigate as technology changes how people work.
A learner might need to respond to an employee's uncertainty, discuss responsible experimentation, navigate competing priorities, or evaluate disagreement about an AI-assisted workflow.
The simulation provides an opportunity to make decisions, experience realistic conversational responses, and reflect on how different approaches affect the interaction.
Afterward, detailed assessments developed by organizational psychologists help learners understand their performance, examine critical moments, identify strengths and development opportunities, and explore recommendations for continued improvement.
This connects the experience of practicing a conversation with more meaningful feedback about the leadership capabilities involved.
It's not about teaching managers to memorize the perfect response.
It's about helping them develop the judgment and communication skills needed to navigate situations where the right response depends on context.
Connecting leadership practice to organizational development
There is also a broader opportunity for organizations investing in AI transformation.
As we've explored in previous articles, Practice Ground combines individual assessments with organizational analytics to help learning leaders better understand development needs.
That perspective can be particularly valuable during organizational change.
Different managers may encounter different challenges as AI adoption progresses. Some may need support communicating expectations, while others may benefit from additional practice navigating uncertainty, evaluating evidence, or addressing employee concerns.
Understanding these development needs can help organizations make more informed decisions about where to focus leadership training.
Of course, simulation-based assessments are one source of developmental information, not a definitive measure of workplace performance or organizational AI readiness.
Their value lies in helping organizations identify areas for further learning and practice.
And when AI adoption creates new demands on managers, that insight can be especially useful.
AI transformation depends on people, not just technology
Organizations will continue to evaluate AI investments through productivity improvements, cost savings, adoption metrics, and business outcomes.
Those measures matter.
But behind the numbers are people making decisions about how work should be done.
Employees are experimenting with new approaches. Managers are evaluating results and establishing expectations. Teams are negotiating changes to workflows, responsibilities, and standards.
Sometimes those conversations will be straightforward. Other times, they'll involve disagreement, uncertainty, or competing interpretations of the evidence.
The organizations best positioned to navigate these challenges will need more than capable technology.
They'll need leaders who can encourage innovation while exercising sound judgment, evaluate evidence without dismissing different perspectives, and guide employees through change.
That's the opportunity behind Practice Ground's Leading Through AI Change training solution.
Because AI adoption isn't just about teaching people to use new tools.
It's about preparing leaders for the conversations those tools create.
