Is your AI strategy teaching employees to hide productivity gains? Dr. Corina Taban on human-centered adoption

Companies want employees to embrace AI, rethink how they work, and deliver major productivity gains. But for many employees, one question remains unanswered: What happens to me if this works?

In the fourth episode of The Agentic Wave, host and Parloa CMO Latané Conant talks with Dr. Corina Taban, CEO of 934 Advisors and organizational behavior researcher focused on the human side of AI adoption. They explore how AI is changing the relationship between companies and employees, why incentives matter, and what leaders can do to increase AI adoption.

Employees want to know what AI means for them

Every employee has a formal contract. But the relationship between a company and its people also depends on a psychological contract: the expectations and perceived promises that never appear in writing.

Employees expect that processes will be fair and that good work will create opportunities. The thought has been that if we invest in the business, the business will invest in our growth.

AI is putting pressure on that bargain. Employees are being asked to learn new tools, often on their own time, redesign how they work, and become more productive, all without a clear picture of what comes next. As Corina puts it, the message many employees hear is: "We'll invest in the business, and your employability is your problem."

Even when leaders haven't connected AI to headcount reductions, employees wonder whether getting too good at AI could automate them out of a job. For Corina, that uncertainty is the adoption challenge.

If an employee saves time with AI, what’s her incentive to tell anyone?

Corina illustrates an incentive problem through Mary, a hypothetical analyst who finds a way to complete a report in a fraction of the time it once took.

Now, Mary has a few options.

She can tell her manager and receive more reports to work on. She can keep the improvement to herself and reclaim the time. Or she can share it broadly and risk giving the company a reason to conclude that it needs fewer analysts.

“Today, the reward system for using AI well is usually either more work or less stability,” Corina says.

To Mary, hiding the gain can feel like the most rational choice.

But there’s still another option. Mary could share what she learned and receive part of the value she created: more flexibility, time back, a raise, or a new opportunity. The company still benefits, but now Mary has a reason to share her productivity gain.

If organizations want employees to discover transformative AI use cases, they have to answer the question employees are already asking: Who benefits when I become dramatically more productive?

When leaders leave a void, fear fills it

The reality is, many leaders don’t know exactly how AI will change every role just yet. Some may hesitate to communicate anything to their employees because they don’t want to make promises they can’t keep.

But silence gives people room to write their own version of the future.

“Ambiguity has a way of making people fill in the gaps with the worst-case scenarios,” Corina says.

Those scenarios aren't hard to come up with. Corina points out that the broader narrative around AI at work has been built on distrust: AI is coming for your job, then your kids' jobs, and AGI is always six months away from “turning us into pets or get[ting] rid of us” altogether.

According to Corina, not knowing what will happen can erode trust as much as layoffs or restructuring. Sometimes, not communicating is worse than sharing bad news.

Give your team time and space to learn AI by doing

When it comes to getting confident in using AI, passive training isn’t enough. Corina points out that adults learn by doing. If people don’t engage with it, it won’t stick. They’ll plan to try it later, and then life gets in the way.

That puts the responsibility on leaders to make room for hands-on practice: dedicated time, support, and resources to try a real use case, see what goes wrong, and try again.

Latané shares how Parloa approached that challenge with an AI day for its marketing team. Instead of telling people to build an agent, leadership asked everyone to arrive with a use case of their choosing. They got a full day to build it, plus a coach on call to help when they got stuck.

As Latané puts it, “You can’t learn to ride a bike at a seminar. It’s the same for AI.”

Reward outcomes, not more AI usage

Hands-on practice works best when it’s aimed at the right problems. Before asking employees to adopt AI, leaders need to get clear on what they're trying to achieve. After a period of "throwing stuff at the wall," Corina says it's time to slow down and define the business value, the strategy, and a realistic timeline.

“You do not need AI for everything,” Corina says. “Let’s sit down with a pen and paper, think about what we’re trying to achieve.”

That clarity helps employees focus on where AI can remove repetitive work, improve an outcome, or make a new process possible, instead of using it everywhere just because they can and are told to.

The same goes for measurement. Tracking prompts, tokens, or logins rewards activity, not outcomes. As Latané jokes, you can consume plenty of tokens organizing your spice drawer.

Build a human-centered AI strategy

Every challenge in this conversation comes back to how people experience AI at work. Corina sums up what it takes to address them in four components of a human-centered AI strategy:

• Psychological safety: Leaders share their own AI experiments, including the failures, so employees feel free to experiment too.

• Support: Employees get dedicated time, relevant training, and help when they get stuck.

• Trust: Leaders explain why the organization is adopting AI and how people fit into the future it's building.

• Incentives: Employees get a meaningful stake in the value they create, so sharing gains beats hiding them.

Corina closes with a reminder: the relationship between employees and employers is built on reciprocity. When people feel they're getting a good deal, they go above and beyond. When they don't, they adjust, whether that means disengaging, resisting AI, or leaving.

Watch the full episode of The Agentic Wave to hear Corina and Latané discuss the psychological contract behind AI adoption, how to build trust with AI agents, and why employees need a meaningful stake in the productivity they create.