How AI Models Are Helping to Understand — and Control — the Brain
SMRTR summary
A groundbreaking study has revealed that AI models can now control high-level brain activity without invasive procedures. Martin Schrimpf, a neuroscientist at the Swiss Federal Institute of Technology Lausanne, led a team that trained an AI to generate sentences capable of activating or suppressing specific neural patterns in readers' brains.
"I thought I could ask neuroscientists how the brain works, and that would help me build better AI," Schrimpf explains. "But I realized there's a huge opportunity in the opposite direction: prototyping ideas in silico and using AI models to explain the brain."
This breakthrough could potentially revolutionize treatments for conditions like depression and dyslexia. However, Schrimpf acknowledges the ethical concerns, stating, "This is an ethical minefield. We need to work with experts on that."
As AI continues to blur the lines between human and machine cognition, Schrimpf remains optimistic about its potential to unlock the mysteries of human intelligence.
SMRTR provides this summary for quick context. The original article belongs to Quanta Magazine.
Read the original article