Lessons from the hacks

Recent cyberattacks exploiting frontier AI models reveal misaligned incentives between fast-growing tech companies and slow-moving governments, highlighting a need for greater transparency and preparedness to manage AI risks effectively.