More than 100 security researchers argued that restricting frontier AI models harms defenders more than attackers and urged the administration to reverse its order.