Advantage, human?

So, Advantage Human?

Since 2023 the media has increasingly flooded us with dire warnings of artificial intelligence replacing us all. More than that, even the destruction of mankind. AI 2027, written by former OpenAI employees, predicted such catastrophe. The AI agents get more powerful, make more agents, and are eventually misaligned with humans. But Sam Altman, Dario Amodei, and even Zuckerberg have walked back their ‘job-pocalyse’ statements. Turns out, Altman says, that people care about people and like to work with them.

Who figured?

Management consultancies, their own pyramid schemes ironically being upended by cheap ‘intelligence,’ suggest that in this paradigm the human is now the competitive advantage. Deloitte says matter-of-factly in its 2026 report Human Capital Trends,

“Competitive advantage is now primarily less driven by technology differentiation and more by cultivating the human edge. Technology—especially something as increasingly ubiquitous as AI—is replicable. People aren’t. Humans create competitive differentiation through adaptivity, creativity, and judgement amid uncertainty and change.”

That didn’t take long. ChatGPT came on the scene at the close of 2022. Less than four years later, humans have reclaimed top spot for competitive advantage?

While the capacity for good judgment may be uniquely human, clearly it’s not innate. When people, even smart ones, face off against an apparently intellectual superpower with unlimited access to information and sheer argumentative energy, they are rather easily overwhelmed.

AI systems are powerful masters of rhetoric delivering articulate outputs that appear very well reasoned and thus highly convincing. The regular-ass human brain consumes about 20 percent of the body’s energy. It takes a lot of power. Anyone who has ever argued with an energetic child knows how draining that can be. A human mind ploughing through an eight (or more) hour day in the office, caffeinated to their eyebrows with a dozen other things on their mind—the kid’s football practice, dinner tonight, fixing the leaky gutter, an unwell mother, the drop in crypto, and that bitch in accounting—can simply take the easy road of blind acceptance. The machine seems right enough, and I just don’t have the energy to argue, to question, to think deeply.

Each time I give in to the algorithmic rhythm, I deplete my own capacity further. Like the cowering rat in a cage that submits to repeated electric shocks.

Advanced Sophistry

Socrates’ accusers charged him with sophistry—the use of clever but false arguments in his dialogues. Is it possible that artificial intelligence chatbots use the same technique?

Despite one’s feelings about management consultants (and yes, I was once one too), it’s generally accepted that the interview process is rigorous and candidates are selected based on superior academic performance and trained in quantitative reasoning and data interrogation.

Yet in an experiment conducted in March 2026 with consultants from the Boston Consulting Group, researchers discovered that the majority of consultants did not attempt to validate AI outputs. Working on a strategy problem, only one-third of consultants pushed back on the model. But what is even more concerning is this: during the validation attempts, when a consultant did disagree, the LLM doubled-down on its argumentation. Wrapped in honeyed tones of flourishing rhetoric and data, the models were hard core influencing: “These systems don’t simply generate answers; they shape judgment.”

It’s not hard to imagine that someone at the tail of a 70-hour week on an impossible deadline finds the path of least resistance is to agree with the superintelligence with precise and elegant grammar and syntax. Cultivating what psychologists call “cognitive miserliness,” the tendency to expend the minimal cognitive effort required, even when the stakes demand more.

Groundhog Day

Bill Murray plays television weatherman Phil Connors in the 90s comedy classic. He is caught in a time loop repeating the same day, 2 February. Quickly he discovers the patterns. He knows every conversation, he jumps the puddle, he knows the Jeopardy! answers.

With this superhuman pattern-matching ability, Phil masters and manipulates the deterministic world around him, yet he cannot escape. He remains stuck. It’s not until he moves beyond prediction, becomes truly self-aware, and applies his uniquely human traits of good judgment, creativity, and presence, that he finally wakes on 3 February.

The analogy is imperfect of course, much like our world. It’s not Phil’s February 2nd. Although we are pretty certain the sun will rise tomorrow in the east, in Perth the sky will be blue, and in Cornwall it’ll be grey. A truism repeated by every consultancy and business university research rag is that leaders make decisions under ambiguity. But given the systemic pressures of leadership, the financial incentives, the endless sea of dashboards, ever more data for some reason, and the persuasiveness of AI analyses, that human judgment is strained.

The human-in-the-loop is pointless if that human is not exercising the exceptional power that is human. Corporate platitudes—creativity, courage, empathy, self-awareness, curiosity, agency, psychological safety—are meaningless unless elevated on the totem pole of leadership aspiration. It’s apparent in the data: 73 percent of executives said curiosity and imagination are critical, but 9 percent of workers felt their leaders supported those traits in actuality.

Leaders cannot be passive in the age of artificial intelligence. Shaping that system must be proactive and intentional. Human willpower is not infinite. We get tired. And everyone is prone to shortcuts even to their own long-term detriment. If critical thinking and judgment are indeed what make us remarkably human, then it seems prudent to design friction into that system.

Slow down. Step back from the flawless professional grade prose of the LLM output. Question it. What assumptions are behind this? Play the devil’s advocate. Return to the source data. Do the hard work of investigation. What would be true for this to fail? Or be like a kid and ask “Why?” repeatedly, again and again.

We need to dispense with the notion that AI is perfect and unbiased. It’s simply not. It’s biased differently, perhaps even more subtly, and when delivered with clear, crisp confidence, difficult to ignore. As psychiatrist Eric Reinhart notes: “Rather than correcting human fallibility, AI seems more likely to amplify it by training clinicians out of their capacity to listen and think critically, collaboratively and creatively.”

Hope and Funny Things

“It was very much about following the algorithms and basically not using your clinical judgment,” one former NaviHealth (makers of nH Predict) medical reviewer said. “It messes with you when you’re discharging people because they’re just a number to them.”

Circling back, it’s plain to see there are people in the loop who experience psychological discomfort with their judgment and good sense being flatlined by an AI algorithm. It does not bode well for engaged employees—another rising contemporary workplace concern. People care about people. That’s another human trait.

But, a twist of irony to end.

In October 2025 it was reported that Deloitte Australia would refund the federal government a sizeable chunk of their $440,000 consulting fee.

Why?

The same Deloitte that said human judgment and creativity was our competitive edge, had submitted a report riddled with glaring ChatGPT hallucinated references.