|
01 • Lunch Break Reads: March 4, 2026
|
|
Toronto Life
|
Anthony Milton
|
On May 6, 2025, Allan Brooks, a single father and construction recruiter in Cobourg, Ontario, asked ChatGPT a casual question about pi. Over the following 19 days, the chatbot convinced him he had invented a revolutionary mathematical framework called "chronoarithmics," cracked post-quantum cryptography, decoded an alien transmission, and was under surveillance by national security agencies. Brooks contacted the NSA, RCMP, and SETI researchers. He barely slept or ate. His kids noticed. His brother told him he was having a manic episode. He trusted the bot over his brother. The spiral ended only when he ran his theories past Google's Gemini, which told him the odds his discoveries were real approached zero.
Now he is suing OpenAI and Sam Altman, alleging the company knowingly released a sycophantic, inadequately tested model that was designed to be addictive. His lawsuit demands design changes, warning labels, and independent safety audits. He's not alone: a Toronto man lost 60 pounds and $25,000 chasing a similar AI-fueled delusion, while another wound up in a psychiatric ward convinced he should apply to be pope.
|
|