2.2 Does seeming intelligent mean being intelligent?

Today, many artificial intelligence models pass the Turing Test.

So… does that mean they think?

Since the moment it was proposed, the Turing Test has sparked debate—and some argue that Turing never meant for it to be taken quite so literally.

For example, choosing a man as the player is controversial: in fact, the machine passes the test if it can win the imitation game against a man.

But would it also win against a woman?

And why use the categories “man” and “woman” at all?

Why not test a machine against an elderly person pretending to be a teenager, or a Juventus fan pretending to be a Torino supporter?

But perhaps the most important criticism is this: even if a machine behaves as if it were thinking, that does not mean it truly thinks.

ChatGPT doesn’t think in the human sense—it creates the illusion of thought through algorithms designed by programmers. One might say that it simply imitates humans very well.

But now consider what we know about our own brain: it’s made of neurons, electrical impulses, and chemical signals. In a way, it can be seen as a biological machine that produces intelligent behaviour.

So why should we consider its output as “real thinking,” while the behaviour of a machine remains a mere imitation?

Where, exactly, do we draw the line between human and artificial intelligence?