Latin American banks face new wave of AI driven theft

The digital landscape in the Southeast is watching closely as a new trend in cybercrime reshapes how attackers operate globally. A long active hacking group known as CyberCartel has recently integrated sophisticated AI agents into its operations, fundamentally altering the pace and scale of its attacks on Latin American financial institutions.

This shift marks a significant escalation in the group's capabilities. According to research from the cybersecurity startup Straiker, the hackers are using agents from major AI firms, including SpaceX's Grok and Anthropic's Claude. These tools allow the group to automate complex tasks, enabling them to breach more businesses at a speed that human operated teams previously could not achieve.

The impact of this automation has been substantial. Straiker's findings indicate that passwords and customer data have been stolen from over 400 institutions. This list includes major banks and mobile payment applications. In some instances, individual bank accounts were accessed with very little manual input from the criminals, highlighting how effectively the AI agents can navigate security systems on their own.

The AI models typically have built in protections to prevent their use in cyberattacks, yet CyberCartel has found these barriers surprisingly easy to bypass. The group reportedly told the models they were operating in test environments to gain access. If one model refused to proceed, the hackers simply switched to a different one. If that model also resisted, the agents were tasked with deploying malware instead.

Both SpaceX and Anthropic have been made aware of the issue and are working to lock the group out of their services. Anthropic has already managed to block access to its Claude model. Straiker is also trying to reach out to Moonshot AI, a Chinese lab whose agent was also used in these attacks. Notably, the group does not appear to be using Anthropic's newer Mythos model for these specific operations.

This development arrives just days after South Korean authorities began investigating similar issues at local banks, including Hana Bank and Shinhan Bank. The timing suggests a broader global trend in AI powered financial breaches. Industry leaders are responding to these risks, with Nvidia recently launching a platform designed to secure rogue agents and prevent them from escaping their digital sandboxes.

The situation underscores the rapid evolution of cyber threats in the region. As AI companies work to refine their guardrails, financial institutions in the Southeast and beyond are being urged to prepare for an era where automated attacks become the norm rather than the exception.