Back to Articles
AI Isn’t Going to End Humanity ... Right? – Podcast

The Guardian

READ

Details

Date Published
14 Sept 2026
Priority Score
3
Australian
No
Created
14 Sept 2026, 08:00 am

Authors (6)

Description

A series of figures in the tech world have been warning about – and even resigning over – the existential threat posed by AI. How worried should we be? With technology editor Robert Booth

Summary

This podcast explores the rising alarm among tech industry figures and developers regarding a self-estimated 10% risk of AI-driven human extinction within the next decade. It examines the shift of AI safety concerns from a niche technical discourse to a central priority for the global political class, citing high-profile resignations and instances of lost control over frontier models. The discussion highlights the tension between rapid capability advancements at companies like OpenAI and Anthropic and the urgent need for governance frameworks to mitigate catastrophic risks.

Body

00:00:0000:00:00AI isn’t going to end humanity ... right? – podcastA series of figures in the tech world have been warning about – and even resigning over – the existential threat posed by AI. How worried should we be? With technology editor Robert BoothTen per cent. That’s the risk, more or less, claimed a high-profile AI developer last week, that the technology wipes humanity out in the next decade.And he was not alone – over the last few months there has been a string of warnings, resignations and even stories of AI companies losing control of their bots.“I think it’s a crossover moment for the AI safety story because this has been for a long time talked about inside the AI world”, says our UK technology editor, Robert Booth, “but now, the general public and the political class have become increasingly concerned about it.”So how worried, asks Annie Kelly, should we be? Photograph: Dado Ruvić/ReutersExplore more on these topicsAI (artificial intelligence)Today in FocusOpenAIAnthropic