Inside the AI Evolution Debate: Allen Dorin’s Insights, the Moravec Paradox & Why We’re Shaping Intelligence, Not Losing to It
📍 By GlobalTimesAI.com
Artificial Intelligence (AI) is transforming our world at a blistering pace. While humans took billions of years to evolve from single-celled organisms to intelligent life, AI has reached astonishing capabilities in just a few decades. This contrast raises both awe and alarm: Is AI really dangerous for humans? Or is our fear rooted in misunderstanding its evolution?
The Misconception: AI as an Instant Threat
One of the loudest fears in tech discourse is that AI might one day replace humans entirely — take over jobs, dominate decision-making, or even rebel.
But here’s what few people talk about:
👉 The danger isn’t AI becoming “too powerful” overnight — it’s us misunderstanding how AI evolves.
Enter the Moravec Paradox
A principle proposed by Hans Moravec and supported by scholars like Allen Dorin (Monash University), the paradox says:
“Hard tasks for humans (like chess) are easy for AI. Easy tasks for humans (like walking or recognizing faces) are hard for AI.”
This paradox flips everything on its head. Playing Go or solving math equations is no sweat for GPT-like systems. But understanding emotions, morality, or body movement? That’s still incredibly hard.
Why?
These abilities took humans millions of years of evolutionary refinement.
Professor Allen Dorin is an academic at Monash University (Australia), known for his research in artificial life, generative systems, and the intersection of biology and AI. His work focuses on how AI can mimic or evolve like natural systems—not just behave intelligently, but emerge, adapt, and co-evolve with environments and humans.
He does not claim AI is dangerous in the way sci-fi movies do. Instead, his research emphasizes:
- Co-evolution: AI develops with human interaction and data.
- Creativity in Machines: Machines can generate art, music, or even living-like behaviors through algorithms.
- Ethics & Emergence: Human-like behavior may emerge from simple coded rules, but without human context, AI remains “mechanical”.
Is AI Dangerous?
Only if misunderstood. The danger lies not in AI becoming “alive”, but in humans thinking it’s smarter than it really is. AI lacks ethics, empathy, or wisdom—unless trained, supervised, and ethically aligned.
Dorin’s work encourages us to see AI as a mirror, not a monster.
Evolution vs. Engineering
Animals like dolphins developed echolocation through natural evolution because they needed it for survival. But it took millennia.
AI, on the other hand, didn’t evolve organically — it’s engineered. It doesn’t learn to “need” skills. Humans train it, feed it with datasets, adjust its weights, and correct its outputs.
“AI is not born smart. It’s trained. By us.”
– Allen Dorin, “Brains, Machines and the Imitation Game” (2016)
The Imitation Game: What Allen Dorin Really Says
In his well-cited research paper, Dorin revisits Alan Turing’s Imitation Game — often misunderstood as the only benchmark for AI intelligence.
According to Dorin:
“The Imitation Game is not about replicating human behavior superficially — it’s about understanding whether a machine can meaningfully interact with us using logic, context, and language.”
Dorin challenges the idea that mimicking humans is the ultimate goal of AI. Instead, he warns against anthropomorphizing AI — giving it human traits it does not actually possess.
AI Mimics, But It Doesn’t Understand
AI doesn’t “know” it’s answering your question or “want” to help you book a flight. It’s mimicking behavior learned from massive amounts of training data.
Just like a toddler learning to walk, AI learns to respond — but without the human sense of agency or moral compass.
AI Isn’t Just Taking Jobs — It’s Creating Them
Instead of fearing mass unemployment, we should recognize that new professions are already emerging:
- ✅ AI Trainers – Teach models what’s right or wrong
- ✅ Prompt Engineers – Design smart prompts for optimal AI results
- ✅ AI Safety Testers – Detect biases, hallucinations, and dangers
- ✅ Ethicists – Define rules and moral guidelines for AI systems
- ✅ Human-AI Collaboration Designers – Build workflows where both work in harmony
Global Impact: A New Era of Evolution
We are no longer just evolving biologically — we are entering a co-evolutionary era, where human decisions shape the growth of synthetic intelligence.
If left unchecked, this could lead to unpredictable outcomes — not because AI “wants” to harm us, but because we didn’t build the right guardrails.
Final Thoughts
This isn’t a sci-fi apocalypse. It’s a historical shift.
A new form of intelligence, built by us, evolving with us.
Next time someone says “AI will replace us”, ask them:
“Who trained it to begin with?”
Because AI didn’t emerge from nature.
It emerged from us.
Sources:
- Allen Dorin, Brains, Machines and the Imitation Game, Monash University
- Moravec, H., Mind Children (1988)
- Turing, A., Computing Machinery and Intelligence (1950)
- Journal of Artificial Intelligence and Simulation
Disclaimer: All information and data presented in this article are collected from reliable sources, research publications, and public statements. The images used are AI-generated and intended for illustrative purposes only.