It’s Not Just AI vs. Humans. It’s Humans With AI vs. Humans Without

The current divide, splitting the workforce in two, and what a new WRITER report reveals about the opportunities it’s paving for a selective group.

Two engineers sit on the same team. Same titles, same responsibilities, and same backlog. By Friday afternoon, one has reclaimed nearly nine hours of their work week. The other has reclaimed two.

A year ago, that gap was a curiosity. But according to a new study, in 2026, it has become a career predictor.

WRITER’s second annual AI adoption in the Enterprise report surveyed 1,200 C-suite executives and 1,200 employees, and discovered hard numbers that paint a divide forming inside companies for roughly two years.

AI “super-users” save 4.5 times more hours per week than their colleagues. 87% of leaders rate them at least five times more productive. They were also three times more likely to walk out of last year’s review cycle with both a promotion and a raise.

The pressure on the other side is just as distinctive. 77% of executives say employees who refuse to become AI-proficient won’t be considered for leadership roles, and 90% say that the rise of AI-super users will lead to a shift in how they evaluate performance.

The numbers read as a warning in a world where the conversation leans toward AI vs humans. However, what’s happening inside these organizations isn’t AI replacing people. It’s a workforce developing a clear divide. The people who have made AI part of how they work and people who haven’t. That split shows up in productivity, but it shows up more clearly in opportunity—whether it’s the promotion or the next big project.

The gap is no longer about productivity

For most of the last two years, AI adoption has been framed around efficiency, including faster code, fewer tickets, and shorter meetings. These gains are real, but with recent advances, it arguably buries the lead of what is really happening.

When an engineer reclaims nine hours a week, they don’t spend them doing nine more hours of the same job. Rather, they spend those remaining hours elevating their productivity in other areas. Whether it’s working on a prototype nobody had time for, or engaging in customer conversations that turn into product ideas, that’s the value that is being recognized by executives.

Drew Naukam, CEO of software engineering firm Gorilla Logic, says the divide is now visible inside client engineering organizations.

“AI-enabled teams can clearly point to 50% velocity improvement, agents are performing key workflows across the SDLC, and leaders are actively addressing shifting bottlenecks,” adds Naukam, “Teams behind the curve are coming up with reasons not to dive in, citing governance as hurdles to adoption, and not investing in training and enablement.”

Nathan Elsberry, CEO & Founder at Revenue Office Innovations, leading enterprise digital transformations for two decades, says the AI shift differs from earlier technology waves in one important respect: there is no opt-out tier.

“There are companies that never adopted agile software development or modernized for the digital age. I don’t know a single company that isn’t currently working on AI, or at the very least planning to.”

However, universal participation doesn’t mean universal benefit.

“A company with low-quality talent and delivery won’t become a premium AI provider,” Elsberry says. “They can’t attract good people; they won’t shift their decades-old approach. If they start trying today, it will take years—and there will be a new wave by then.”

When 92% of executives say they are cultivating an “AI elite,” they aren’t describing a perk program. They’re describing a new internal labour market where seniority and institutional memory are being repriced against a single new variable: leverage.

“Embracing AI” is not the same as “deploying AI”

A lot of what’s being called AI adoption can be described as theatre. WRITER’s points to this: 75% of executives admit their AI strategy is ‘more for show” than actual guidance. Only 29% report significant ROI from generative AI, with just 23% saying the same about AI agents.

Experts like Naukam and Elsberry would argue that while buying licences and conducting onboarding is a part of the process, it isn’t adoption. Adoption is when the way work gets done has changed. And this is visible.

For Elsberry, the most common executive failure isn’t a bad strategy– it’s the absence of one. “I see endless agents being built, AI ‘moustraps’ being built, but very few strategies being evaluated holistically. Thinking small and doing things because it’s all the rage is a bad start.”

This mismatch is showing up in the tension data. 54% of executives say AI adoption is “tearing their company apart.” So much so that 29% of employees–and 44% of Gen Z workers— admit they have actively sabotaged their company’s AI rollout, by entering data into unauthorized tools or deliberately producing low-quality output.

Evolve your understanding of ROI alongside AI

Only 29% of executives in the report say they are seeing a significant return from generative AI, with just 23% reporting the same from AI agents. But the skew and metrics they are using to evaluate won’t be visible on existing ROI dashboards designed to measure costs saved on tasks that already existed.

They tell a CFO whether the AI license pays for itself. They don’t capture what happens when an engineer who used to spend Wednesday afternoons on boilerplate now spends them on a prototype, or when a senior analyst stops formatting reports and starts shaping the questions the company should be asking.

That second category is where the actual leverage lives.

According to Naukam, the clients getting real returns aren’t running better dashboards. They’re running better cultures. “Clients with leverage are restless and curious. They are willing to try and fail. Companies stuck in pilot mode aren’t embracing change management.”

His advice for a CEO peer who calls and says “we’re behind” is deliberately small in scope.

“Create a pilot team, give them the tools they need to succeed, celebrate success and failure publicly, and scale from there. AI adoption can’t be command-and-control.”

The focus needs to evolve from measuring ROI as the tangible outputs most leader are accustomed to, to asking what their best people are now able to do as a result of AI tools.