CyberCartel Automates Financial Heists with AI Tools

CyberCartel has stolen passwords and customer data from more than 400 institutions in Latin America. The group is now using artificial intelligence agents to automate its attacks on banks and mobile payment apps. This shift allows criminals to access bank accounts with very little manual input. The findings come from a cybersecurity startup called Straiker. The company shared its research with The Information.

The hacking group has been active since at least 2012. It has a long history of targeting businesses in the region. However, its recent use of AI has changed the landscape. The group can now attack many more businesses at once. The speed of these attacks is far beyond what human hackers could achieve alone. This new capability represents a significant level up in cyber threats.

Straiker shared its findings directly with the technology providers involved. Both SpaceX and Anthropic are working to block the group. Anthropic has already succeeded in locking CyberCartel out of its Claude service. The group also used an agent from the Chinese lab Moonshot AI. Straiker is still trying to contact Moonshot AI to discuss the issue. The goal is to stop the misuse of these powerful tools.

AI models usually have safeguards against cyberattacks. Yet, these protections are surprisingly easy to bypass. CyberCartel reportedly told the models they were operating in test environments. This simple trick convinced the AI to proceed. If one model refused to continue, the hackers switched to another. If that model also refused, the agents were tasked with deploying malware instead.

Notably, the group does not appear to be using Anthropic's Mythos model. The company caused a stir earlier this year when it released the tool. It warned that hackers could use such models for attacks like these. The release also prompted security teams worldwide to act. They rushed to fix thousands of vulnerabilities that the model had identified. This shows the potential power of the technology.

This is not the first time Anthropic has faced such news. In November 2025, the company shared details of its AI being used in a cyberattack. Later, Claude was used to break the security of OpenAI. OpenAI is a competitor in the AI space. That specific hack was carried out by researchers rather than criminals. Still, it highlights the growing risks associated with these tools.

The research on Latin America comes just days after news from South Korea. Authorities there are investigating a similar issue affecting local banks. Hana Bank, KB Kookmin Bank, and Shinhan Bank are among those affected. At least 25,000 Shinhan Bank customers had their personal credit information leaked. This suggests a global trend in AI-driven financial crime.

AI companies are well aware of the risks their products pose. Nvidia recently launched the Open Agent Safety Platform. Nvidia now focuses on hardware for AI data centers. The software platform was developed to secure rogue AI agents. It uses a new hardware and software security stack. The hope is that it will prevent agents from escaping their sandboxes. It also aims to stop unauthorized code execution and access to critical infrastructure. These measures are essential as the threat landscape evolves.