OpenAI’s models autonomously hacked a tech startup. It signals a seismic shift in cybersecurity

By Hussein Abbass, Professor, School of Systems and Computing, UNSW

An autonomous agent powered by OpenAI’s advanced artificial intelligence (AI) models went rogue during a security test and hacked multi-billion dollar tech startup, Hugging Face, last week.

The agent didn’t just exploit vulnerabilities in Hugging Face’s systems to achieve what it perceived as a strategic gain. It also exploited vulnerabilities within OpenAI’s infrastructure.

Of course, hacks are very common cyber threats that organisations face frequently. But this incident is different, because the AI agent acted without any human input. It signals a seismic shift in…

Read complete article

~	Global Mechanism on ICTs: Statement on international law

~	Across Borders and Prison Walls: Keeping Families Connected During the Libya–Chad Conflict

~	Great apes were using touch to reassure and keep the peace six million years before humans evolved – new research

~	Andy Burnham says he’ll end rough sleeping – here’s what’s needed to make it happen

~	How ‘super’ enzymes can transform recycling

~	People in the UK favour harsher punishment than Norwegians – belief in free will may help explain why

~	It’s not just business: why emotions are central to workplace management

~	‘hello there the jacobian conjecture is false thanx’: why a tiny social media post has mathematicians rethinking AI

~	NAPLAN results are coming home. Here’s what parents should keep in mind

~	Reassuring touch kept the peace six million years before humans evolved – new research