Researchers suggest medical AI systems could be vulnerable to adversarial attacks
A small team of medical researchers from Harvard University and MIT has published a Policy Forum piece in the journal Science suggesting that future medical AI systems could be vulnerable to adversarial attacks. They point ...