The AI efficiency paradox: What enterprise leaders must know before costs outweigh productivity benefits

We also gave you a sneak preview of the survey results, so if you’ve not checked it out, head here.

Now we want to go a little deeper into some of the ideas the data has raised–and you’ll hear from our experts and industry-leading technology partners (who speak in detail in the report) about how to begin addressing these challenges.

Generative AI tools are everywhere, and while it’s hard to imagine a future workplace without them, many workers are looking back with rose-tinted glasses. Almost two-thirds (65%) of knowledge workers say they regularly feel nostalgic about what work was like before the widespread adoption of AI. There’s a clear reason behind what we call ‘pre-AI nostalgia’: people feel the heart and soul of their work, the human bit, is slipping away.

While many agree that AI has made work more efficient, what about creativity? What about individual expertise that’s taken years to accrue? And what about personal experience–the special something people bring to the workplace?

But let’s be clear–nostalgia doesn't mean regression. Our data shows that people feel less valued sometimes, but in reality, their value has shot up. Those workers who really know their stuff and can spot where AI agents have gone wrong are even more essential to business success today.

For AI strategies to succeed, though, you need to support those workers by preserving what's important: their autonomy, craftsmanship, and identity–and then investing in AI where it delivers the most value.

AI is designed to lighten the load, yet our data shows that there's actually a hidden productivity cost to using these tools, which we refer to in the report as the ‘verification tax’. That's all the time people spend checking, correcting, and validating machine-generated content. Rather than producing it themselves, workers are now effectively marking AI's homework.

And the numbers speak for themselves: 42% of workers spend more time verifying AI output than they save using it, and 52% regularly correct AI-generated work from colleagues.

Clearly, this dynamic isn't sustainable, and leaders need to eliminate the verification tax on their employees. This starts by leveraging AI functionality in the project management tools teams are already using, automating the more mundane, low-risk administrative tasks that make work feel repetitive and less meaningful. With that taken care of, specialists can refocus their energy on high-value strategic and creative work–and begin eliminating the verification tax.

Our findings reveal a new cultural dynamic arising from AI use at work, in which people feel they are pitted against it. This ‘human vs. machine’ phenomenon has emerged as they feel the speed, volume, and quality of human work is compared to that of AI.

Half of workers (50%) feel their performance is now directly or indirectly compared to AI-generated output, believing their professional value is being assessed through a lens that pits human intuition against algorithmic efficiency.

The data also shows that people are using AI to keep up with their workload, stay one step ahead, and work faster and more efficiently. Underpinning this is a fear of obsolescence–a growing fear that their days in the workplace are numbered. And our findings show that senior leaders aren't immune. In fact, they're the most uneasy about the possibility of their role being needed in the future.

This represents a real risk to successful digital transformation; if your team feels they are losing ‘against’ AI, adoption and engagement will stall.

The data highlights that the mandate is now clear for business leaders–now is the time to focus on building a culture that encourages openness, two-way dialogue, and a thoughtful approach to changing workplace dynamics. By moving beyond the mindset of human vs. machine and recalibrating performance benchmarks to reward the complex strategy work and nuance that only humans can provide, businesses can turn the tide of anxiety around AI.

The narrative surrounding AI has shifted from technical feasibility to a more complex cultural friction: the ‘human vs. machine’ dynamic. Currently, Half of workers (50%) feel their performance is now directly or indirectly compared to AI-generated output, believing their professional value is being assessed through a lens that pits human intuition against algorithmic efficiency.

People are also using AI to keep up with their workload, stay one step ahead of colleagues, be faster and more efficient, and produce higher-quality work. Underpinning it is a fear of obsolescence, that their days in the workplace are numbered. And senior leaders aren't immune. In fact, they're the most uneasy about the possibility that their role may be needed in the future.

For leaders, this represents a significant risk to digital transformation; if your team feels they are losing against AI, adoption will stall. The mandate now is to move beyond the "vs." and recalibrate productivity benchmarks to reward the high-level strategy and nuanced oversight that only your human capital can provide.

With pre-AI nostalgia, the verification tax, and a fear of obsolescence in play, it’s no surprise that some people are looking for a way out. According to our research, 33% of knowledge workers are now considering changing industries entirely to escape from the shadow of AI, while others are weighing up an earlier retirement. This white-collar exodus could imply that organisations may face a loss of knowledge and experience if these problems remain unaddressed.

But that’s not the full picture. For nearly three-quarters of knowledge workers (74%), the answer is not to flee, but to actively learn new skills to stay relevant. What’s going to make the biggest difference to which side of the fence people land? Not the technology itself, but the way workplaces support and communicate with their people–and the opportunities they provide to upskill.

If any of the challenges outlined above resonate, the fuller findings and expert commentary in our report offer a detailed evidence base to support your approach to AI.

Download our report, The human cost of AI transformation, to read our complete findings, analysis from our experts, and a path to help you build an AI technology strategy that delivers efficiency without undermining the people that make it work.