Recently, I've found myself wondering what role artificial intelligence might play in this area. Not because I believe AI is creating psychopaths. What interests me is whether some of the ways we interact with AI could gradually influence how we process empathy, guilt, and accountability.

We need to teach children to interrogate everything. Not with anxiety, but with curiosity. To hold up a piece of information and turn it slowly in the light. To understand that a deepfake isn’t just a technical trick: it’s a weapon, and someone made it for a reason.

To ask not just ‘what does this say?’ but ‘what does this want from me?’

And we need to model this for them.

If the reasoning process becomes opaque even to those who are responsible for its consequences, accountability shifts from something lived to something theoretical.

Leaders may still sign off decisions, yet their capacity to interrogate the underlying thinking diminishes if they have no experiential sense of how that thinking was formed.

If you can’t reconstruct how a decision was formed, what assumptions were accepted, what alternatives were rejected, and where human judgement intervened, you don’t really have a decision at all. You have an artefact: a thing that exists, but one that can’t defend itself. In regulated sectors, that’s a compliance issue waiting to happen. In unregulated ones, it’s a reputational risk that just hasn’t found its trigger yet.

If we don’t actively teach people how to question, how to verify, how to tolerate being wrong, and how to think independently of algorithms, we risk creating a society that functions efficiently but thinks very little.

And that’s the quiet fear running through all of these comments. Not that AI will become too intelligent. But that we’ll stop becoming intelligent ourselves.