Weeks after the US government restricted foreign access to Anthropic’s most advanced artificial intelligence models, intelligence agencies from the Five Eyes alliance have issued a rare joint warning, saying powerful AI systems capable of carrying out sophisticated cyber attacks could emerge within months.

The alliance, which comprises Australia, Canada, New Zealand, the United Kingdom and the United States, said frontier AI models are advancing much faster than previously anticipated and could fundamentally reshape both offensive and defensive cyber capabilities.

Agencies Say AI Will Change Cyber Security Landscape

In the statement issued late on Monday Sydney time, the Five Eyes agencies acknowledged that artificial intelligence has the potential to strengthen cyber defences but cautioned that the same technology could also make attacks faster, larger and more complex.

“Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months,” the agencies said.

The alliance warned that cyber security should no longer be treated solely as a technical issue managed by information technology teams.

“In this environment, cyber resilience is integral to advancing business continuity, market confidence, and long-term value,” the statement said.

“A whole-of-organisation and whole-of-society response is required.”

According to the agencies, cyber risk has become a leadership issue and should be regarded as a core business concern.

Concerns Grow Over Advanced AI Systems

Although the statement did not name any specific companies or models, it comes amid growing concerns among governments over AI systems capable of identifying weaknesses in computer networks.

Much of the recent attention has focused on Anthropic’s latest models, Claude Fable 5 and Mythos 5.

Fable 5 has been described as a more widely accessible version of Mythos, which was introduced earlier this year. Mythos is designed to identify vulnerabilities in cyber systems and has drawn scrutiny over concerns that its capabilities could be misused.

Citing advice from national security authorities, the US government recently ordered Anthropic to suspend access to both models for foreign nationals, including users in India.

Reports Link Restrictions to National Security Concerns

A report by Wired said concerns surrounding South Korean telecom operator SK Telecom may have played a role in Washington’s decision.

According to the report, the White House asked Anthropic to revoke SK Telecom’s access to Claude Mythos days before the models were taken offline.

US officials were reportedly concerned about what they described as the company’s alleged links to China.

The report further stated that Amazon researchers had identified methods to bypass certain safeguards built into Fable 5, the public version of Mythos released on June 9.

Researchers argued that such weaknesses could potentially expose some of the advanced cyber capabilities embedded in Mythos.

Anthropic and independent cyber security experts reportedly maintained that the risks identified were not unique to Claude models and reflected broader challenges facing the development of advanced AI systems.

The Five Eyes warning underscores growing concerns among governments and intelligence agencies about the pace of AI development and its potential impact on global cyber security.

Long or Short, get news the way you like. No ads. No redirections. Download Newspin and Stay Alert, The CSR Journal Mobile app, for fast, crisp, clean updates!

App Store –  https://apps.apple.com/in/app/newspin/id6746449540 

Google Play Store – https://play.google.com/store/apps/details?id=com.inventifweb.newspin&pcampaignid=web_share