Available March 12, 2026

Episode 017: Season 2, The Algorithmic Mirror – What AI Reveals About How We Actually Think, Decide, and Deny

DURATION: 32 minutes
HOST: Kevin Novak
SHOW: The Human Factor Podcast

COLD OPEN

I want to start with something that happened at a client I was advising last year. They had deployed an AI system to assist with talent evaluation, promotions, and internal promotion decisions. Within three months, the head of HR approached me while I was onsite for a meeting. She was visibly shaken. She said, “Kevin, the AI is biased. It keeps recommending the same profiles, the same demographics, the employees for promotions.” I asked her to show me the training data. We spent two hours going through it. And when we were done, she sat back and said something I will never forget. She said, “It’s not biased. It’s accurate. It learned exactly what we’ve been doing for the last fifteen years. We just never had to look at it all at once before.”

That moment captures everything we’re going to explore today. The discomfort people feel around AI is rarely about the technology. It’s about what the technology reveals about us.

INTRODUCTION

Welcome to The Human Factor Podcast. The show that explores the intersection of Humanity, Technology, and Transformation along with the psychology behind transformation success.

In our last episode, we explored the middle management trap, the five psychological burdens that organizations place on their most critical change agents without acknowledgment or support. We examined the translation burden, the absorption burden, the identity burden, the loyalty burden, and the accountability burden. One of the themes running through that episode was how middle managers become mirrors for the gap between what organizations say they value and what they actually reward. Today, we’re going to take that mirror metaphor much further. Because there is another mirror now operating in virtually every single organization, and it is far more unforgiving than any manager’s quiet frustration.

That mirror is artificial intelligence.

I wrote about this extensively in my recent newsletter, “The Algorithmic Mirror: What AI Reveals About How We Actually Think and Decide.” Today I want to go deeper, drawing on research published through early 2026, connecting this to the frameworks we’ve built across this season, and exploring what I believe is the most psychologically important question leaders face right now: not whether AI belongs in your organization, but whether you are prepared to see what it reflects back.

SEGMENT 1: THE MIRROR NO ONE ASKED FOR

AI is almost always framed as a disruptive force. Something that replaces human judgment, automates expertise, and accelerates decisions beyond our control.

That framing dominates boardrooms, media coverage, and vendor presentations. But it misses what makes AI genuinely unsettling. The discomfort most organizations and individuals feel around AI is not really about technology or the fear of being replaced by it. It is about the exposure that results. The very core of human thinking, human faults, human biases and human interpretation are brought to the forefront, exposed, and made visible for everyone to see.

AI does not just do work differently than humans. It reflects how work has always been done, only more clearly, more consistently, and without the social camouflage humans rely on to soften or hide uncomfortable truths.

Let me ground this in research. A 2025 study from the University of Washington, presented at the AAAI/ACM Conference on AI, Ethics and Society, tested 528 participants across sixteen different job roles. What they found was striking. When AI systems exhibited racial bias in hiring recommendations, human decision makers mirrored those biases, selecting candidates in line with the AI’s recommendations. But without AI input, or with neutral AI, choices were unbiased or at least in the eyes of humans. The lead researcher put it plainly: unless bias is obvious, people are perfectly willing to accept the AI’s biases as their own.

Now think about what that means. Humans feeds AI the information on which it bases its decisions. The AI learned human patterns. Then humans absorbed those reflected patterns and acted on them, often unconsciously. This is not a story about flawed machines. This is a story about human patterns are being made visible at a scale and speed we were never prepared for or perhaps never wanted to admit and reveal.

A comprehensive 2025 review published in Frontiers in Big Data mapped the entire landscape of algorithmic bias and categorized it into four families. Historical and representational bias, which reflects the values embedded in training data. Selection and measurement bias, which reflects what organizations choose to measure and what they ignore. Algorithmic and optimization bias, which reflects the objectives humans defined for the system. And feedback and emergent bias, which reflects the reinforcing loops between human behavior and machine output. Every single category traces back to human decisions, human-curated data and human input. Not one of them is a purely technical artifact, although we would like to think that is the case.

The Uncomfortable Inversion

Here is what makes this so psychologically difficult. For decades, organizations have relied on the opacity of human judgment as a feature, not a bug. Decisions could be explained through experience, intuition, relationships, all of the soft language that makes subjective choices feel authoritative. AI removes that opacity. It makes the logic visible. And when the logic is visible, the inconsistencies become undeniable.

Research published in Management Science in 2025 found something both reassuring and deeply uncomfortable. Even fairness-unaware machine learning algorithms, systems never designed to be fair, can reduce bias in human decisions.

If an algorithm that was never designed to be fair produces less biased outcomes than a human process, what does that tell us about the human process? It tells us that the judgment we trusted was never as objective as we believed. We defined our own version of the truth and accepted subjectivity as objectivity.

This is precisely what I explored in the revised version of my book, The Truth About Transformation, Leading in the Age of AI, when examining how humans shape and distort the systems they build. What we are seeing now in the research confirms what that analysis anticipated: the technology is not the problem. The human system feeding and using the technology is the problem.

SEGMENT 2: WHAT THE MIRROR ACTUALLY SHOWS

So, when organizations and individuals look honestly at algorithmic outputs, three things consistently become visible. And each one challenges a different aspect of organizational self-image.

The Mirror Reveals What Gets Rewarded

The mirror routinely reveals what gets rewards, not what is written in mission statements, but what actually advances careers, earns recognition, and avoids friction.

If an algorithm prioritizes volume over quality or speed over reflection, it is because the system learned those preferences from human behavior. Consider how we think and construct the prompts we input, what language we use, how that language can be evaluated, and the inferences AI makes about what we are really seeking.

MIT Sloan Management Review found that twenty-six percent of respondents reported getting meaningful personal value from AI while noting that their organizations derived little or no value.

That gap between individual benefit and organizational return is not a technology failure. It is a mirror that reflects what organizations and their leaders actually reward versus what they say they reward. When individuals extract value but the organization does not, you are seeing a system where personal optimization has been prioritized over collective alignment. AI did not create that dynamic. It made it measurable.

The Mirror Exposes Where Judgment Was Already Thin

Many roles assumed to require deep expertise are, in practice, governed by heuristics, shortcuts, and unspoken rules. AI does not eliminate judgment in these cases. It reveals how thin it already was.

The California Management Review published a 2025 study that should be deeply uncomfortable for anyone who believes human judgment is inherently superior.

When AI was deliberately debiased, it outperformed human decision-making in board candidate selection by enabling a shift from fast, unconscious processing to slow, conscious evaluation.

The researchers titled their paper “Slow Thinking Fast,” a deliberate inversion of Daniel Kahneman’s framework.

What they demonstrated is that AI can force the kind of deliberate reasoning that humans claim to exercise but rarely do. It is not that human judgment is bad. It is that human judgment is often faster, more intuitive, and more socially acceptable than it is accurate.

I wrote about this dynamic in my newsletter on professional identity crisis, “When Your Expertise Becomes Obsolete.” The paradox remains: the employees organizations rely on most, the go-to problem solvers, the institutional knowledge keepers, often become the people most psychologically threatened by change. Not because they oppose progress, but because systematic change threatens the very expertise that defines their professional identity.

This also connects directly to what we explored in Season 2, Episode 1, on the identity crisis of expertise.

AI does not just threaten what people do.

It threatens the story people tell themselves and others about who they are.

The Mirror Surfaces Inconsistency

Our new algorithmic mirror surfaces inconsistencies we likely didn’t see before. Humans routinely make exceptions, override rules, and rationalize deviations. We are always seeking to sway others to our values and ways of thinking. When AI applies the same logic consistently, those inconsistencies become visible. Not subtly visible. They become impossible to ignore.

Harvard Business Review reported in January 2026 that bias in AI is not just baked into training data. It is shaped by the broader ecosystem of human-AI interaction. The way people engage with AI, through their thinking, questions, interpretations, and decisions, significantly shapes how these systems behave.

Organizations do not just train algorithms on their data. They train them on their culture. And culture includes every exception, every override, every unspoken compromise.

The issue is not then that AI lacks judgment. It is then that it forces organizations and individuals to confront how selectively they have been using theirs.

SEGMENT 3: WHY WE RESIST THE REFLECTION

So why do we so very strongly resist the reflection? Why do we not want to look into the mirror?

Resistance to AI is often framed as caution, ethics, risk management or even policy. Sometimes those concerns are entirely legitimate. We know hallucinations occur. We see deepfakes that appear disturbingly realistic. We question whether our prompts actually produce what we intended. But often, the stated concerns mask something much deeper: identity threat.

The Psychology of AI Resistance

Lets take a deeper look at the psychology of AI resistance.

For many leaders and experts, authority has long rested on opaque judgment, the ability to make decisions others cannot fully explain or replicate.

AI challenges that mystique.

When an algorithm produces similar or better outcomes using transparent logic, it raises an unsettling question: what, exactly, has expertise been hiding? Not hiding deliberately.

But hiding structurally, because the opacity of human judgment has never before been challenged at this scale.

A foundational study published in Electronic Markets identified three central predictors of AI identity threat in the workplace. First, changes to work, the structural alteration of what someone does day to day. Second, loss of status position, the erosion of hierarchical authority. And third, what the researchers call AI identity, the degree to which professionals see AI as challenging who they are, not just what they do.

That third dimension is the one most organizations completely miss. They address the functional threat, retraining, upskilling, and process redesign. They mandate the use of AI and gauge performance on such. However, they rarely address the existential one.

A peer-reviewed 2025 study published in MDPI Systems went further, identifying two distinct dimensions of AI anxiety. Anticipatory anxiety, driven by fears of future disruption, the worry about what might happen.

And annihilation anxiety, reflecting existential concerns about human identity and autonomy, the fear that one’s professional self might cease to matter.

The language itself, annihilation anxiety, captures something that leaders must take seriously.

This is not resistance born of stubbornness.

This is resistance born of an existential threat to professional selfhood.

Connecting the Season 2 Arc

Think about what we have built across this season. In Episode 1, we explored the identity crisis of expertise, how professionals whose value rests on specialized knowledge experience AI as a fundamental threat to who they are. In Episode 2, we examined the contagion effect, how those identity fears do not stay contained within individuals but spread through emotional contagion, amplifying at each organizational layer. In Episode 3, we explored the middle management trap, how the people most responsible for translating change into reality are carrying psychological burdens the organization refuses to acknowledge.

Now in this episode, we are seeing the mechanism that accelerates all of those dynamics.

AI acts as a mirror that makes the gap between espoused values and actual behavior visible at unprecedented speed and scale.

That visibility intensifies identity threat.

That identity threat feeds emotional contagion.

And the people caught in the middle of that cascade, your middle managers, absorb the full force of the organizational anxiety without the authority or support to address it.

This is not a series of separate problems. It is one interconnected psychological system.

And AI is the catalyst that makes the entire system visible and very real simultaneously.

The Role of Denial

In my newsletter on invisible friction, I explored how strategies falter not because people are obstinate but because they fear loss. They disengage not because they are lazy but because they feel disconnected from meaning.

AI amplifies this dynamic because it introduces a new form of loss: the loss of narrative control. When an algorithm reveals that your decision-making was less objective than you believed, that your expertise was thinner than you presented, that your organization rewards behaviors it publicly condemns, you lose the ability to maintain the story you have been telling yourself and others.

The most common response to that loss is not adaptation. It is denial. Blaming the tool for revealing uncomfortable truths. Demanding human override without examining the reflected human bias. Rejecting outputs instead of interrogating underlying assumptions.

SEGMENT 4: THE RECURSIVE TRAP AND THE SCALE OF DENIAL (8 minutes)

In this type of situation, here is where AI bias becomes genuinely dangerous. AI bias is not linear. It is recursive.

Organizations train AI on their historical decisions. AI reflects those patterns back. Humans absorb those reflections and act on them, often unconsciously.

And the cycle repeats, each iteration reinforcing what came before. Research in Nature Human Behaviour has shown that this amplification is exponential rather than linear, with humans often unaware of AI’s influence, making them more susceptible to the recursive pattern.

Meanwhile, transparency is declining. As organizations become more dependent on AI systems, the biases reflected back become harder to detect, harder to attribute, and harder to correct.

You end up in a situation where the organization is looking into a mirror that is growing more distorted with each reflection, and no one can see the distortion because they have normalized the image.

The Scale of Failure

The scale of organizational denial about this dynamic is staggering. MIT’s 2025 report, “The GenAI Divide,” based on interviews with 150 leaders, surveys of 350 employees, and analysis of 300 AI deployments, found that ninety-five percent of corporate AI projects fail to create measurable value. Ninety-five percent. Meanwhile, ninety-one percent of CIOs cite culture as the primary impediment to AI adoption, versus only nine percent citing technology.

Read those numbers together. Leaders largely agree that the technology works. The human systems do not. Or more precisely, the human systems within organizations refuse to engage with what the technology reveals. They are looking into the algorithmic mirror and choosing not to see.

The McKinsey Global Survey on AI, published in March 2025, found that the single biggest factor in achieving measurable returns from AI is the redesign of workflows, a fundamentally human and cultural undertaking. High-performing organizations were three times more likely to have redesigned workflows around AI capabilities rather than simply layering AI on top of existing processes. The value of AI comes from rewiring how organizations run. That is not a technology project. That is a transformation project. And transformation, as I have argued consistently throughout this podcast, my books, my newsletter, and my research, lives or dies in the human system.

Why Traditional AI Implementation Fails

Most organizations unfortunately, approach AI implementation as a technology deployment. They focus on selecting tools, training users, and building governance frameworks. All necessary. All insufficient. Because they address the functional dimension while ignoring the psychological one.

Until implementation approaches address the story people tell themselves and shape, that identity narrative, they will continue to fail at the rate MIT documented.

I wrote about this cognitive dimension in my newsletter on the mental overload of modern leadership. Leaders today are already operating at the limits of their cognitive architecture, processing across multiple complex systems simultaneously, while their brains evolved for sequential processing. Adding AI to this environment without addressing the fundamental mismatch between cognitive capacity and role demands does not solve the overload problem.

It amplifies it.

Our new mirro reflects faster than leaders can process what they are seeing.

SEGMENT 5: LEARNING TO LOOK

So what do organizations do with this? How do leaders move from denial to engagement with what the algorithmic mirror reveals?

Reframe the Question

Let’s reframe the question. Most AI conversations focus on control. How do we govern it? How do we constrain it? How do we limit risk? Those are necessary questions, but they are incomplete. A more uncomfortable, and ultimately more useful, question is this: what is AI revealing about who we are and how we already operate?

That reframing changes everything. Instead of asking “how do we prevent AI bias,” you ask “what human biases is AI making visible?” Instead of asking “how do we maintain human override,” you ask “what are we overriding and why?” Instead of asking “how do we build trust in AI,” you ask “why did we trust our existing processes when the evidence suggests they were never as objective as we believed?”

Five Practices for Engaging the Mirror

Let me outline five practices that leaders can begin implementing to move from denial to productive engagement with what AI reveals.

First, conduct an algorithmic audit of values alignment. Take your organization’s stated values and compare them against what your AI systems have learned from your actual behavior. Where the algorithm’s outputs diverge from your stated values, you are looking at the gap between aspiration and reality. That gap is not an AI problem. It is a leadership opportunity.

Second, create psychological safety around AI-revealed truths. When AI surfaces uncomfortable patterns, the organizational response matters enormously. If people are punished for what AI reveals, you guarantee deeper denial. If the revelations are treated as organizational learning opportunities, you create the conditions for genuine change. This connects directly to the psychological safety research from my transformation psychology series on building psychological safety during transformation.

Third, address identity threat directly. Stop pretending that AI anxiety is simply a training problem. Acknowledge that for many professionals, AI represents an existential challenge to their sense of professional self. Create structured conversations where people can express that threat without being dismissed as resistant or told to simply “get on board.” This is what we discussed in Episode 1.

Fourth, break the recursive loop. Establish regular intervals where you examine what AI has learned from your organization and ask whether those learned patterns reflect the organization you want to be or the organization you have been. This requires the kind of honest organizational self-assessment that most companies avoid because it is psychologically uncomfortable. But avoiding it guarantees that the bias feedback loop continues to compound.

Fifth, redesign workflows before deploying AI. The McKinsey data is clear. Organizations that redesign workflows around AI achieve measurably better outcomes. But workflow redesign is not a technical exercise. It is a psychological one. It requires people to let go of processes that have defined their professional identity, to accept that the way things have always been done may not be the way they should continue. That is transformation work. And it requires the kind of human-centered approach that the Human Factor Method was built to address.

CLOSING

Let me bring this together. AI is not forcing organizations to change who they are. It is forcing them to see who they have been. For leaders willing to look honestly, that reflection can be transformative. For those who refuse, AI will remain an external threat, something to resist, regulate, or blame.

The technology is not the mirror. It is the light. And once something is visible, it cannot be unseen.

Across this season, we have been building a comprehensive picture of the human psychology of transformation. The identity crisis that change triggers. The emotional contagion that spreads that crisis through organizations. The structural trap that crushes the people in the middle. And now, the algorithmic mirror that makes all of it visible at a scale and speed that organizational denial cannot contain.

In our next episode, we are going to explore what happens when all of these psychological dynamics collide with something even more deeply embedded: organizational culture. I will be joined by a senior executive who has led cultural transformation at the enterprise level, and together we will examine what I am calling the organizational immune system, the cultural antibodies that organizations develop over years of success that attack anything perceived as foreign to the established order. We will draw on Edgar Schein’s foundational work on organizational culture and connect it to what we explored in Season 1, Episode 11 on organizational drift. Because understanding why organizations can intellectually grasp everything we have discussed this season and still fail to change requires understanding how culture operates below conscious awareness, how it creates invisible resistance that no strategy deck or AI implementation can overcome without first being diagnosed.

If you have not yet subscribed to The Human Factor Podcast, find us on Apple Podcasts, Spotify, YouTube, Amazon Music or anywhere you watch or listen to podcasts. You can explore more at 2040digital.com, and I encourage you to read the full newsletter that inspired today’s episode, “The Algorithmic Mirror,” for the complete research and analysis.

Until next time, remember: the question is not whether AI belongs in your organization. The question is whether you are prepared to see what it reflects back. And if you are, what are you willing to do about what you see?

This is The Human Factor Podcast. I’m Kevin Novak. Thanks for watching or listening.

END OF EPISODE

🎵

Upcoming: Episode 018: THE ORGANIZATIONAL IMMUNE SYSTEM – WHEN CULTURE ATTACKS WHAT IT DOESN’T RECOGNIZE

Learn about the organizational immune system, the invisible cultural defense mechanism that kills more transformations than bad technology ever could.

Season 2 Launched on February 20, 2026

More Episodes Coming Soon

View Main Podcast Page →