Only 12% of Workers Say AI Has Transformed How Work Gets Done

Only 12% of U.S. workers at organizations that have implemented AI strongly agree the technology has transformed how work gets done, according to Gallup's State of the Global Workplace 2026 report. That stark figure sits right next to a much cheerier one: 65% of those same workers say AI has had some degree of positive impact on their individual productivity.

The gap between those two numbers is the story. Darren Person, EVP and chief product and digital officer at Cengage, argues that AI's early gains have centered on helping individuals complete existing work faster — and that the bigger opportunity lies elsewhere. The real question is how what one employee learns can change how an entire team, function, or organization works.

The problem is easy to picture. Today, two employees sitting directly across from each other can have dramatically different experiences with the same AI tool. One might use the technology to write an email. The other may have discovered a way to redesign a process that saves significant time and resources. If those use cases stay with the individuals, the organization never captures their full value.

Person's core argument: as AI becomes increasingly embedded across organizations, access to the technology itself is less likely to create sustainable advantage. The differentiator will be how quickly organizations can turn what employees learn into shared knowledge, redesigned workflows, and better ways of working.

What does leadership actually change here?

Employees are closest to day-to-day work, so they're often the ones uncovering the most practical ways AI can improve processes or solve problems differently. They understand where existing processes create friction.

Walmart offers one concrete example, as reported by Business Insider. Rather than treating employees only as end users of AI, the company has involved them in shaping AI applications. One logistics manager helped create an AI process designed to improve truck efficiency and reduce empty trailers. The insight didn't come from someone studying the workflow from a distance. It came from someone close enough to the work to see what needed to change.

Person says leaders create the conditions for that type of discovery by:

• Establishing the problems worth solving
• Setting the boundaries for experimentation
• Defining what a better outcome looks like

Employees then have room to figure out how AI can change work.

At Cengage, Person's company, AI was introduced in phases. "Change Champions" across the company experimented with AI, identified meaningful use cases, and shared what they learned with their teams. As access expanded, employees had examples, practices, and colleagues they could learn from — rather than asking every team to start from scratch. The result: nearly 90% of employees with access are now actively using AI.

Can structured programs spread learning faster?

One employee finding a better way to perform a task creates individual productivity. Dozens of employees learning from that discovery, adapting it, and improving it creates organizational impact — that's Person's framing. The difference compounds across departments.

Canva ran a version of this experiment last year. The design software company gave 5,000 employees a dedicated week — an AI Discovery Week — to learn and experiment with the technology and explore how it could improve their work. The value extends beyond what employees accomplished during that week, Person argues. The dedicated space created an opportunity for what they learned to spread across the entire organization.

Traditional training still plays a role, Person notes. It helps employees build a shared foundation and understand how to use AI responsibly. But training alone can't keep pace with a technology changing this quickly. The most useful applications often emerge as employees try AI against the real challenges in their own work. Organizations need to make it easy for those employees to share what they've learned, so a promising idea becomes a starting point for others.

Cengage's Change Champions illustrate the loop: employees with subject matter expertise learned by experimenting, then brought practical examples back to their teams.

Why managers are the missing link

A culture of shared learning requires more than encouragement. Employees need time and support to test new ideas and share what they learn — without expecting every experiment to deliver an immediate return.

The data backs this up. Microsoft's 2026 Work Trend Index found that when managers created psychological safety around experimentation, employees reported up to 20 points higher AI readiness and value.

Senior leaders can set the expectation that AI experimentation matters, but employees experience culture primarily through their managers. That puts managers in an increasingly important position. Person identifies what they can actually do:

• Protect time for experimentation
• Help teams identify worthwhile problems
• Set expectations for quality and responsible use
• Make sharing lessons part of the team's regular work

Managers are also close enough to distinguish between an interesting demo and a change that genuinely improves a workflow. Person calls them the bridge between individual experimentation and organizational learning.

What does this change for the average company?

The next phase of AI adoption is an organizational learning challenge, not a procurement one. Models will change. Tools will change. What seemed like a leading-edge use case six months ago may quickly become commonplace — that pace makes one-off retraining programs obsolete.

Organizations cannot retrain the entire workforce from scratch every time the technology shifts, Person writes. They need a system that helps employees discover, test, share, and build on better ways of working — continuously.

His closing bet: the organizations that ultimately create the most value from AI won't be the ones with the best tools. They'll be the ones that build the capacity to continually learn what the technology makes possible, and turn those individual discoveries into better ways of working across the whole organization. For practitioners, that means the 65%-versus-12% gap is the scoreboard to watch next.