Huang rejected Geoffrey Hinton's 10%+ catastrophic AI risk estimate as unscientific, saying such predictions lack research support.
"Just because it comes from a scientist doesn't make it scientific," Huang said, arguing that such predictions were not supported by research.