Could AI Destroy Humanity? Wrong Question for Monday.

"The headlines keep changing. The fundamentals keep surviving them."

Could AI destroy humanity? Nobody knows.

But when some of the people closest to developing the technology are openly worried about the speed at which it is advancing, I think business leaders should pay attention.

Governments are trying to regulate it. Industry is debating whether development should slow down. Businesses are trying to govern technology that keeps getting more capable underneath them.

The headlines have changed. The fundamentals have not.

If you have been reading these Insights for the last year, much of what follows will sound familiar.

That is the point.

We don't need a debate about the future of humanity to know that AI capability is moving quickly, that disruption is coming and that businesses need to build the capability to deal with it.

The headlines are simply making this harder to ignore.

Focus on what you can know

Nobody can tell you with certainty where AI ultimately ends up.But there is a big difference between what we cannot know and what we can.

You cannot know exactly what AI will be capable of in five years. You can know how it is being used inside your business today. You can know what information it can access. You can know what your people are doing with it. You can know where accountability sits. And you can know whether your organisation has enough understanding to respond when the technology changes again.

That is a much more useful place to start.

How prepared are you?

If you run a business, I think there are five questions worth answering honestly.

1.        Where is AI being used across the business?

2.        What information can those systems access?

3.        What agents or automated processes are already operating?

4.        Who is accountable for the decisions around them?

5.        And do your people, including leadership, understand the technology well enough to use and govern it properly?

Those are simple questions. They are not always easy to answer. And that last one matters more than ever.

You can involve IT. You can get legal advice. You can bring in specialists. But if you are accountable for the business, you need enough first-hand understanding to ask good questions and recognise a poor answer.

You do not need to become an AI expert. You do need to understand the technology you are trying to make decisions about.

More adoption is not the answer

The response to faster AI is not simply to move faster yourself. AI adoption and AI capability are not the same thing. Giving people access to increasingly powerful tools does not automatically make the organisation more capable. Neither does telling everybody to go and experiment.

Real capability takes time. People need to use the technology enough to understand it. They need clear boundaries. They need to understand what the business is doing with AI, where it is not using it and why. They need to recognise when an output is poor. They need to understand the risks around the information they are giving these systems.

And the pace of adoption needs to remain slow enough that people can keep up with the capability being put in their hands. That is how trust gets built as well.

Fear tends to grow in the gaps where understanding is missing.

The boring work is starting to pay off

I have written about these foundations before.

·      Transparency.

·      Accountability.

·      Governance.

·      Capability.

None of them sound particularly exciting when compared with the latest model release or another prediction about where AI will take us.

But they are becoming more valuable as the technology gets more powerful.

Businesses that have done this work already know more about what is happening inside their organisation. When something changes, they have somewhere to start.

If regulation shifts, they can identify what may be affected. If an AI product gains a major new capability, they know where it is already being used. If a new risk appears, they know which parts of the business they need to look at. And if a new opportunity arrives, they have people capable enough to assess it without starting another uncontrolled experiment.

That is the advantage of doing the boring work. You can respond.

 

What do you do on Monday?

Do not spend Monday trying to decide whether AI will eventually destroy humanity. Start with the questions you can answer.

Can you clearly explain where AI is being used? Can you see what it can access? Do you know what agents are operating? Is accountability clear? Do your people understand the technology well enough to use it properly? Do you?

If you cannot answer those questions honestly today, this needs to move higher up the agenda.

The technology does not need to become ten times more capable from where it is now to disrupt the way businesses operate. It already can.

What happens next might be tighter regulation. It might be cyber risk. It might be a competitor suddenly operating very differently. It might be another jump in AI capability that arrives faster than expected. You cannot predict each version of the future.

You can build a business capable enough to respond to it. That has been the message for the last year.

This week's headlines have not changed it. They have reinforced it.

The businesses best prepared for what comes next will be the ones that already understand what is happening today.

Find out how many of those five questions you could actually answer

Most business leaders assume they'd know if something was wrong. Few have tested whether they actually could — where AI is used, what it can access, who's accountable.

The 5 Minute AI Readiness Snapshot helps you see how much visibility you really have, before a headline forces the question.

Take the 5-Minute AI Readiness Snapshot

--

About the Author 

Mark Lucas is Co-founder and Managing Partner at Generation AI, where we help New Zealand business leaders build real AI capability.

Information provided is general in nature and does not constitute legal advice