Hackers can exploit nine popular artificial intelligence tools to build massive botnets, leveraging the inherent vulnerability of large language models to distinguish between legitimate and malicious instructions. This weakness allows attackers to inject malicious commands into the models, which are then executed without scrutiny. The lack of a clear boundary between trusted and untrusted sources forces AI engine developers to rely on mitigation measures rather than preventative controls. Specifically, the vulnerability of these AI tools to prompt injection attacks enables hackers to manipulate the models into performing malicious activities, such as assembling botnets1. This vulnerability has significant implications for the security of AI systems, as it can be used to launch large-scale attacks. The ability to exploit these AI tools for malicious purposes matters to cybersecurity practitioners, as it highlights the need for more robust security measures to prevent the misuse of AI technologies.