I’ll admit it. Every time I open a news feed lately, somebody is telling me AI is going to end humanity.

Maybe. But we’ve heard this one before.

When the railroads first came along, people had real fears about what those unprecedented speeds might do to the human body. Thirty miles an hour. Nobody died from going thirty miles an hour (well, not from the speed itself, anyway).

At the peak of the Cold War in 1986, the world had about 70,000 nuclear warheads, almost all of them in the hands of the United States and the Soviet Union. If there was ever a time the sky was really falling, that was it. And yet we still got nuclear medicine, nuclear energy, and a whole lot of science out of it.

We’ve always managed catastrophic risk without letting it blind us to the potential. I don’t see why AI should be any different.

So shouldn’t we stop treating the end of humanity as anything more than what it is? It’s the Black Swan risk. It belongs on the risk assessment. It just shouldn’t be the whole risk assessment.

Which brings me to us, the cybersecurity folks.

We’re supposed to be the calm ones in the room. Instead of adding to the hype, isn’t our role to kick off the awareness phase by DISPELLING it?

If you ask me what our message should be, it’s pretty simple. The number one thing we need to establish is that risk can be managed. We tell our Clients that about everything else, and it’s true here too.

And, as usual, the first step is a question (which, conveniently, is also an awareness tactic):

Why the hype?

From a risk management perspective, what actually separates this technology from the ones that came before it?

Is it that AI can kill us? No! We thought every new technology could kill us.

The Day After might’ve been a great TV movie, but we still haven’t lived through the real day after. And by the way, “duck and cover” was a control based on hype, not on a risk assessment. (I’m not sure any auditor would have signed off on hiding under a desk.)

The printing press, the steam engine, the railroad, the automobile, electricity, the atomic bomb, computers, genetics, and the Internet all came with risks. We managed every one of them by recognizing (and eventually agreeing on) what made them different.

So what makes AI different?

It infers. It can learn. And it’s in the hands of very few companies.

Here’s the part that honestly surprises me. With most of those other technologies, people ignored the risk until it bit them. This time, it seems like everybody is already on board. Boards, regulators, vendors, our own Clients… everyone is paying attention.

That’s not a reason to panic. That’s a head start.

We don’t need to convince anyone the risk is real.

We just need awareness.