Could AI really take control of humanity? With ex-NASA scientist Will Marshall

About this episode

Space entrepreneur and former NASA scientist, Will Marshall, believes we're on the cusp of "the most profound choice humanity has ever made". He tells Richard about the extraordinary potential of AI – and the risks it could pose to humanity. Will, who is co-founder and chief executive of space company Planet, believes there could be a 50% or greater risk of humans losing control of AI if we continue on our current trajectory. He suggests we could be approaching recursive self-improvement – when machines independently improve their own code and capabilities. And he discusses what could happen if AI begins to develop at an accelerating pace, and the potential "one-way door" of handing over power. You can email your questions to Richard and Yalda at theworld@sky.uk. You can also watch our podcasts on our YouTube channel here.