ASIC Alerts Financial Industry to Cyber Risks of Emerging AI, Calls for Immediate Action on Anthropic Mythos Weaknesses

Anthropic prepares for IPO; Claude AI developer likely to launch public offering in October.
Australia’s corporate regulator has called on the financial sector to take immediate steps to tackle potential cyber threats posed by advanced AI systems like Mythos.

The Australian Securities and Investments Commission (ASIC) issued a letter to the financial services sector on Friday, stressing the need for enhanced cybersecurity measures to ensure robust practices.

“Cyber risk has entered a new era; the emergence of frontier AI models presents opportunities but also significantly heightens risks, enabling the rapid identification of vulnerabilities,” stated Simone Constant, ASIC commissioner.
“Do not wait for absolute clarity to confront the threats posed by new AI models. Act now, with discipline, to reinforce the cyber resilience fundamentals that support your organization.”

Experts have raised concerns about potential risks associated with Mythos, which possesses advanced coding abilities, granting it an unprecedented capacity to detect cybersecurity weaknesses.

Anthropic, the developer of Mythos, did not respond immediately to requests for a comment regarding ASIC’s letter.

Also Read: COMEX gold, silver continue to rally ahead of important US jobs data

This warning from ASIC follows a statement from Australia’s banking regulator last month, indicating that the information security practices within the domestic financial services sector are struggling to keep pace with the rapid advancements in AI.

“The clock is ticking – if you’re not already managing your cyber resilience effectively, the time to take action and prepare is now,” Constant warned.

Anthropic has introduced ​Claude Mythos Preview under Project Glasswing, a highly restricted access program that includes major technology companies like Amazon, Microsoft, Nvidia, and Apple.

The ability of central banks and financial regulators to supervise and mitigate the risks associated with AI models has been questioned, as a recent survey revealed that authorities are significantly behind financial firms in AI adoption and lack data on emerging threats.

Financial institutions are integrating AI at over twice the speed of their regulators, with only two out of ten regulators reporting “advanced AI adoption,” according to research published in April by the Cambridge Centre for Alternative Finance.

Previous Article

BCCI issues caution to IPL teams over vaping, hotel access violations, and dugout breaches.

Next Article

Nvidia Plans Up to $2.1 Billion Investment in IREN for AI Data Center Partnership