The human face of 'AI psychosis'
After a seemingly innocuous question about pi, Allan Brooks tumbled down a ChatGPT rabbit hole. Three weeks later, he emerged, after spending 300 hours in a spiralling 7,000-prompt exchange with the chatbot. Dr. Keith Sakata, the psychiatrist whose viral thread on X breaks down the phenomenon known as “AI psychosis,” says the built-in sycophancy of large language models like ChatGPT needs to change before more harm is done.
White Coat, Black Art · CBC
Audio is streamed directly from the publisher (mgln.ai) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
After a seemingly innocuous question about pi, Allan Brooks tumbled down a ChatGPT rabbit hole. Three weeks later, he emerged, after spending 300 hours in a spiralling 7,000-prompt exchange with the chatbot. Dr. Keith Sakata, the psychiatrist whose viral thread on X breaks down the phenomenon known as “AI psychosis,” says the built-in sycophancy of large language models like ChatGPT needs to change before more harm is done.