Listen now (47 min) | Within the coming decades, artificial general intelligence (AGI) may surpass human capabilities at a wide range of important tasks. We outline a case for expecting that, without substantial effort to prevent it, AGIs could learn to pursue goals which are very undesirable (in other words, misaligned) from a human perspective. We argue that AGIs trained i…

On my holiday last December 2022, I decided to delve into why people are so alarmed about the potential disruption caused by artificial intelligence in the near future. Being a researcher on conciousness and morality, I thought venturing out further into why aligning AGI to human intentions has not taken off yet would make for an interesting project. Du…