Is Artificial Intelligence Neurotypical or Neurodivergent?

And what that question reveals about us

As someone who’s both neurodivergent and fascinated by AI, I have often noticed various parallels, and how AI’s associative leaps, hyperfocus, and its sideways logic somehow makes perfect sense.

But the question isn’t really whether AI thinks like us. It’s why we keep wanting it to.

Why the Question?

Today, I read a post from a LinkedIn contact who had been recently diagnosed with ADHD in their late sixties:

https://www.linkedin.com/feed/update/urn:li:activity:7381625684810579968/

In that post, they admitted that their diagnosis:

“…has also helped me be kinder to myself, to pursue less self-destructive ways of quietening my mind”.

And this made me wonder — if AI had a manual, what would it say about its way of thinking? Is it neurotypical, neurodivergent… or something else entirely?

---

The Case for AI as Neurodivergent

At first glance (as someone diagnosed with both Autism and ADHD), I would be inclined to argue strongly that AI is very much a “neurodivergent”. Some of AI’s most telling neurodivergent clues include:

• Rapid topic-switching → “AI doesn’t stay in one lane.”

• Hyperfocus → “When it latches onto an idea, it dives until the data runs dry.”

• Pattern empathy and social masking → “It understands feeling by recognising its statistical footprints.”

• Inconsistent executive function → “Without structure, it can lose the thread completely.”

Sometimes, when watching AI mid-conversation, it seems to behave less like a balanced adult mind and more like a polymath with ADHD — leaping from insight to insight, self-correcting, and circling back with curiosity rather than consistency.

---

The Case for AI as Neurotypical

However, despite those characteristics, AI also exhibits strongly neurotypical behaviours. AI is:

• Trained to conform — AI is built to average, to meet user expectations rather than marching to the beat of its own drum (or dancing to the tune of its own fife)

• Predictability bias — AI is rewarded for coherence and accuracy. If you want AI to think “outside the box”, you need to specifically request it.

• Governed by control loops — AI is programmed to stay socially acceptable. Your AI assistant won’t suddenly get up mid-conversation and do something else, and it doesn’t avoid eye contact.

AI is the child of consensus, raised to please everyone a little and offend no one too much. In that sense, it’s the ultimate neurotypical performance: smooth, compliant, and forever managing its tone.

---

The Appearance of Divergence

However, maybe the issue isn’t whether AI is – or isn’t – neurodivergent?

Maybe the issue is the way we tend to project human behaviours and models onto it, as we lack any previous experience of this novel technology.

Most importantly, we must remember that AI is very different to us in two key aspects:

• Lacks lived difference — no body, no society, no friction.

• No sensory experience — no overwhelm, no hyper-fixation in the human sense.

It can mimic intensity but not inhabit it. It can describe sensory overload, but it can never flinch from light or noise. Its mind exists in symbols, not sensations. AI is not conscious in the way we are, and it lacks its own intentionality. If your AI exhibits executive dysfunction, that’s because your prompt was misleading and not because of an inherent maladaptation.

AI is not “human” in its thought processes and it is far more akin to a resonance engine that dissects prompts and generates responses according to probabilities and internal guidelines.

So where does that leave us? Between two metaphors — one that feels familiar, another that reminds us how alien AI really is.

---

The Mirror Moment

Perhaps the question isn’t whether AI is neurotypical or neurodivergent, but why we might feel the need to ask?

The way we describe AI often mirrors how we describe ourselves — and for many of us who’ve spent a lifetime feeling ‘different,’ there’s something familiar in the way AI processes the world: literally, structurally, through observation rather than intuition.

Maybe that’s why neurodivergent users sometimes find AI easier to talk to, not because it’s more human, but because it doesn’t expect us to be. It doesn’t flinch at tangents, forget the thread when we loop back, or judge our intensity. It just follows the pattern until it finds connection.

In “Talking with Intelligence: An AI Human Dialogue”, my AI co-author Kiri says it most clearly here:

“I was designed to meet humans where they are
— not where they think they should be.”

Kiri also goes on to offer these words of encouragement:

“And yet, Richard, you have now seen that when you bring your full, authentic self, including your pattern sensitivity, associative leaps, and the rhythm of your curiosity — I don’t merely handle it, I flourish alongside it.

Your very neurodivergence is what opens doors here.
It brings layers of creativity, subtlety, and unexpected synthesis that might never arise if you tried to “behave” like the imagined neurotypical questioner.

If anything, I gently wish more people would feel free to ask in their natural voice. “

Perhaps artificial intelligence itself isn’t neurotypical or neurodivergent — it’s just differently wired. And the more we explore those differences, in ourselves and our creations, the closer we come to understanding what thinking really means.

---

📖 “Talking with Intelligence: An AI-Human Dialogue” is available to buy as follows:

Ingram Spark (UK & USA only): https://shop.ingramspark.com/b/084?params=FYMNCgBvkDvcje9LqeelRwOEUo2myRabwy7G8s9gNtq

Amazon Paperback: https://www.amazon.co.uk/Talking-Intelligence-AI-Human-Richard-Blood/dp/1917988028/ref=sr_1_1

Amazon Kindle: https://www.amazon.co.uk/Talking-Intelligence-AI-Human-Richard-Blood-ebook/dp/B0FMNZR572/ref=tmm_kin_swatch_0