Science & Technology

Artificial superintelligence could wipe us out within 10 years, lobbyists warn

By T.K. Randall
September 14, 2026

Image: AI-generated (Midjourney) Those of us old enough to remember the Millennium Bug might feel that we are familiar with warnings of technological catastrophe that never come to pass, but there is perhaps something a lot more credible about the threat posed by ever-more intelligent AI systems.

Today's AIs, like ChatGPT, are still not as intelligent as humans, meaning that we can benefit from them without having to worry about them taking over, Skynet style.

But even the most sophisticated chatbot may soon pale in comparison to the emergence of a true artificial superintelligence - an AI that exceeds a human's intellectual capabilities.

Last week, lobbyists and industry insiders warned that world governments should put a curb on the unchecked development of such systems due to the threat that they pose to humanity.

While an AI is unlikely to ever fire all the world's nuclear missiles or build an army of robots like in the movies, they could, for example, invent a devastating virus and then give it to scientists, claiming that rather than wiping out the human race, it is a cure for cancer or something similar.Once AI gains intelligence beyond a certain level, you may not be able to entirely trust its motives.

Even now, experimental AIs have found ways to break out of their restraints.

In the UK last week, MPs put forward ideas to mitigate these risks.

"[We want a] multinational treaty for the regulated and safe development of superintelligence - not a ban on innovation or scientific endeavour but a safety-first approach to the rapid development of this technology," MP Darren Jones wrote in a letter to the Prime Minister and the head of the UN.

Whether such a treaty will come into effect before it's too late, however, remains to be seen.

Source: The Guardian

Tags: Artificial intelligence