Featured Image Credit: Donato Fasano / Contributor via Getty
Several leading intelligence agencies from powerful nations around the world have come together in a rare joint statement, outlining the potential for 'devastating' AI attacks that could be just months away following the US government's decision to block the release of Anthrophic's Claude Fable and Mythic.
While the conversations surrounding AI have largely been focused on its potential environmental damage, impact on the job market, or ethical concerns, cybersecurity experts are now rushing to warn everyone about the dangers that the rapidly evolving technology has to our worlds.
This has been bubbling under the surface for a while now, with leading figures in the AI industry outlining similar warnings in recent times, yet the development of frontier models has seemingly exceeded expectations, thus presenting a far more immediate threat.
Responding to this seemingly imminent risk, intelligence agencies from Australia, Canada, New Zealand, the United Kingdom, and the United States – otherwise known as the Five Eyes (FVEY) – has called on global leaders to "act now," as the dangers will manifest itself in a matter of months, not years.
As reported by The Guardian, a statement issued on Monday indicated that while AI has the potential to enhance our cybersecurity defense systems – which has been displayed following the release of Anthropic's exclusive tool, Claude Mythos, before it was banned – it concurrently "accelerates the speed, scale, and sophistication of cyber threats."
Advert
AI and national security expert Olivia Shen, who works at the University of Sydney's United States Studies Centre, indicated that "we have to anticipate that the next Mythos or the next Fable is just around the corner.
"We can only see what's been released but there could be other models being developed by the likes of China," she asserts, "or other states and other actors and companies, that are just as advanced."
This new warning urges people to be aware of the 'lower barrier' that AI creates for bad actors, allowing them to carry out attacks with greater ease and speed thanks to newly developed AI tools.
The Five Eyes' statement also points towards the need for societal intervention, which echoes recent arguments made by major tech leaders at the G7 Summit earlier this month.
OpenAI CEO Sam Altman was joined by Google DeepMind and Anthrophic bosses Demis Hassabis and Dario Amodei in demanding that governments take over the direction of AI development, as leaving all that power in the hands of a few is incredibly dangerous.
"A whole-of-organization and whole-of-society response is required," the Five Eyes' statement continued, declaring that "cyber risk can no longer be treated as a purely technical issue. This is a core business risk and leadership responsibility."
It's impossible to know exactly what an AI-driven cyber attack would be able to achieve as the technology is constantly (and rapidly) evolving, yet Sam Altman believes that it would have the power to shake the world.
If something like Mythos were to fall into the wrong hands then it could effectively unlock the cybersecurity protections of some of the world's leading companies and governments, providing bad actors will the ability to spot the same exploits but take advantage of them instead of patching them over.
Additionally, this same sentiment can be applied to many of the most positive AI advancements, as while major strives in health and medical research would provide potentially life-saving treatments, the same tools could be equally used to develop bioweapons and launch terrorist attacks.
That's why so many intelligence agencies and even the development labs themselves are calling so urgently for government intervention, as it's an issue that's not only far bigger than a simple piece of tech but one that is likely just around the corner.
