SYDNEY, April 30 - Australia's financial system regulator, the Australian Prudential Regulation Authority (APRA), warned Thursday that the nation's banks are not keeping pace with AI developments. In a letter to financial institutions, APRA said frontier AI systems like Anthropic's Mythos could empower bad actors to identify vulnerabilities faster, leading to larger and more frequent cyber attacks.
APRA noted that most banks' information security practices are struggling to match the rapid evolution of AI. The regulator specifically cautioned that Anthropic's Claude Mythos, which possesses high-level coding capabilities, could significantly increase the probability, speed, and scale of cyber attacks. Anthropic has not yet responded to requests for comment.
The Australian government is reportedly working with software providers, including Anthropic, to address potential cybersecurity vulnerabilities. APRA's industry consultation found that banks are overly reliant on vendor presentations and summaries without fully considering AI-related risks. Many boards still lack the technical literacy needed to provide effective oversight on AI risks.
Separately, S&P Global said Thursday that AI will affect the credit standing of Asia Pacific financial institutions over the next one to five years, though most banks have large technology budgets to mitigate negative impacts.