AI beyond efficiency: humans and AI

The corporate world is in an AI arms race. Faced with pressure to ‘do more with less,’ companies are rushing to integrate artificial intelligence into their operations – automating workflows, cutting costs, and scaling decisions.

On paper, it sounds like progress. In practice, it often looks like Klarna: the fintech giant replaced 700 employees with AI, only to see its customer experience unravel and its reputation take a hit. The lesson? Automation without alignment is not innovation – it’s fragility in disguise.

What many executives fail to realise is that AI is not just a technical upgrade – it’s a cultural shift. The prevailing mindset treats AI as a silver bullet: plug it in, turn the crank, and enjoy higher margins. But this confuses implementation with adoption. Implementation is tactical; adoption is strategic. The former installs software. The latter reshapes thinking, incentives, and workflows.

Most organisations are chasing efficiency, but the real prize is excellence – the ability to adapt, learn, and innovate continuously. And excellence doesn’t come from better algorithms. It comes from rethinking how people and machines can co-evolve.

The efficiency delusion

The current orthodoxy assumes that more automation equals more value. Companies like Duolingo, for instance, replaced human translators with AI to lower costs, only to face backlash over diminished learning quality. What this reveals is a deeper flaw: efficiency is not a strategy; it’s an outcome. Treating it as the primary goal of AI flattens complexity and narrows ambition.

Here’s the contrarian truth: AI’s biggest opportunity is not to replace humans, but to liberate them. To shift the focus from role execution to creative problem-solving. To enable entirely new kinds of value that weren’t possible before. That doesn’t happen by cutting headcount. It happens by reimagining work itself.

The cultural bottleneck

True AI adoption is a cultural operating system. It rests on a few pillars, including psychological safety, the concept of balancing comfort and discomfort for employees to take well calibrated risks.

In 2012, Google’s Project Aristotle tried to figure out what drives effective teamwork, and found that the top predictor of team performance wasn’t IQ or experience – it was the freedom to fail without fear. In an AI context, this means giving people room to experiment with the tech, critique it, and shape its use. Without this, adoption turns into compliance – and compliance kills creativity.

AI also won’t succeed if it’s imposed. It needs to be co-created, with bottom-up participation, not top-down mandates. When employees are invited to explore AI tools, pilot new workflows, and give feedback, the technology stops being a threat and starts becoming a partner. This approach treats frontline workers as innovators, not just implementers.

But equally, pretending AI won’t change roles is not kindness – it’s dishonesty. The organisations that will thrive are those willing to say: ‘Yes, this will change your job. Let’s shape what that looks like – together’.

The emotional aspect

Culture is built on trust. And today, that trust is broken.

According to Behave’s recent data, 61% of executives view AI as essential, but a massive 68% of employees feel unprepared to work with it. This disconnect isn’t just a communication issue, it’s a strategic risk. If people don’t trust the process, they won’t adopt the tools. And without widespread adoption, AI becomes shelfware – or worse, a source of internal resentment.

The fix isn’t more training. It’s a total recalibration of what AI is for; not just speed or savings, but new forms of value, co-created by humans and AI. To truly unlock AI’s potential, we must reject the binary framing of AI vs jobs, and instead ask about the new capabilities that emerge when humans and AI work together.

This question doesn’t just create new workflows; it creates new business models, new customer experiences and internal contracts between people and technology.

The path forward for AI

Efficiency is easy to sell. But it’s also easy to copy. Innovation, by contrast, is slow, messy, and deeply human. The companies that thrive won’t be those that automate the fastest – but those that adopt AI the deepest, building cultures where AI becomes a tool for amplification, not attrition.

So let’s use AI to elevate judgment, not eliminate it. Design roles that emphasise collaboration, not substitution, and measure success not by cost savings, but by capability expansion.

In the race to embrace AI, don’t just ask what can be automated. Ask what’s worth augmenting.

First published in The Business Reporter.