Headlines:
Deep Dive:
Uh oh – AI is now really, really good at hacking. Earlier this month, the leading AI company Anthropic declared it would hold back from releasing its latest model, named “Mythos”, because the model could find novel ways to hack computer systems. In one instance, Mythos found a bug that's been undetected for 27 years in a fundamental piece of software.
In this episode of The Briefing, Natarsha Belling is joined by Dr. Suelette Dreyfus, Senior Lecturer in the School of Computing and Information Systems at The University of Melbourne.
See omnystudio.com/listener for privacy information.