Âé¶¹ÊÓÆµ

Skip to content

Opinion: Is AI making us smarter or just less human?

We’re racing to automate everything from health care to hiring, but are we losing what makes us human?
just-breath
The future will not be won by logic alone, but by our ability to remember what it means to be human.

The real challenge today isn’t technological — it’s moral.

We are building smarter systems but forgetting what makes us human. As artificial intelligence expands its reach, we risk repeating an old mistake: valuing efficiency over empathy and the collective over the individual. The real threat is not that machines will outthink us — it’s that we will forget how to feel.

The story of the birth control pill illustrates this. A technological triumph, yes, but also a turning point that redefined intimacy and relationships. It gave individuals control over reproduction but turned love into logistics and romance into scheduling. A fix for biology disrupted the emotional fabric of human connection. It was liberating, yet in some ways dehumanizing — an early sign that even our most empowering tools carry unintended social costs.

That same pattern plays out in governance, neuroscience, religion and now in AI. Technology often solves one problem while creating another, especially when it ignores emotional and relational consequences. Our systems grow faster and smarter, but often less humane.

Consider the “tragedy of the commons,” used to explain how individual self-interest can deplete shared resources. In Canada, we’ve seen this in overfished Atlantic cod stocks and overburdened emergency rooms, where rational individual actions create collective harm. But a deeper tragedy emerges when systems serve averages instead of people. Uniformity becomes a form of oppression.

This logic underpins far too much modern policymaking. Whether in education, health care or housing, we are governed by one-size-fits-all models that leave people behind. Wait-lists, formula-driven benefits and automated eligibility systems may look efficient, but they often treat real human needs as exceptions rather than the rule. The most vulnerable pay the price: those who don’t quite fit the system’s definition of “deserving.”

Government has a sacred responsibility: to protect the commons without crushing the individual. This isn’t a policy debate; it’s the work of civilization. And it requires leaders who recognize that equity isn’t about mathematical fairness, but moral discernment.

We see the same tension inside ourselves. Neuroscience shows that the brain is split between logic and emotion. This internal duality plays out in our institutions: law versus mercy, efficiency versus equity. When one side dominates, dysfunction follows.

Religions have long grappled with this same dilemma. The Ten Commandments are fixed but their application allows for grace. In Buddhism, the Middle Way balances competing truths. Indigenous teachings in Canada emphasize relationships between people, land, ancestors and future generations. These traditions remind us: systems must serve people, not the other way around.

Now comes artificial intelligence, the most powerful system we’ve ever created. In Canada, AI is already influencing decisions in health care, immigration and hiring. Credit-scoring algorithms, predictive policing models and automated triage systems are no longer fringe experiments — they’re here. Unless we learn from past mistakes, AI will scale our blind spots. If we train it only to optimize, it will fail to humanize.

We cannot outsource judgment to machines trained only on precedent and pattern. What they lack is what we often neglect: compassion, nuance, moral courage.

True intelligence — human or artificial — requires more than computation. It requires comprehension. It must recognize when to act with empathy, not just precision.

The rise and fall of civilizations often hinge on their ability to balance structure with soul. Ours will be no different. If we continue to build systems that erase the individual and treat compassion as a flaw, we will pay a high price.

The test isn’t whether AI can outsmart us. It’s whether we can out care it. The future will not be won by logic alone, but by our ability to remember what it means to be human.

Dr. Perry Kinkaide is a visionary leader and change agent. Since retiring in 2001, he has served as an advisor and director for various organizations and founded the Alberta Council of Technologies Society in 2005. Previously, he held leadership roles at KPMG Consulting and the Alberta Government. He holds a BA from Colgate University and an MSc and PhD in Brain Research from the University of Alberta.

©

 

The commentaries offered on Âé¶¹ÊÓÆµ.ca are intended to provide thought-provoking material for our readers. The opinions expressed are those of the authors. Contributors' articles or letters do not necessarily reflect the opinion of any Âé¶¹ÊÓÆµ.ca staff.

push icon
Be the first to read breaking stories. Enable push notifications on your device. Disable anytime.
No thanks