Will AI really kill us all?

With headlines full of apocalyptic predictions about AI, Aleks Krotoski and Kevin Fong ask: is this just more hype, or do these technologies pose genuine risks now and in the future?

Ever since a researcher at AI lab Anthropic went public with concerns about the technology he was helping to create, fears about artificial intelligence have returned to the public consciousness. From AI enabling terrorists to develop biological weapons, to killer robots, warnings about catastrophic risks are becoming increasingly common. And with reports of seemingly "rogue" AI agents exploiting vulnerabilities in businesses and government systems, those fears can appear to have some foundation.

So are these threats real? How close are they? And what is being done to protect us?

Aleks and Kevin hear from BBC Reith Lecturer Professor Stuart Russell, Director of the Center for Human-Compatible AI, who believes the risks are serious but avoidable. They also speak to cybersecurity expert Bruce Schneier, who explains both the immediate threats AI poses to our digital infrastructure and the measures being taken to defend against them.

Presenters: Aleks Krotoski and Kevin Fong
Producer: Peter McManus
Research: Elizabeth Ann Duffy and Ciaran Richter

A BBC Audio Scotland production for Radio 4.