Prompt Injection Vulnerability in PandasAI by GetPanda

CVE-2024-12366

What is CVE-2024-12366?

The PandasAI framework leverages an interactive prompt function susceptible to prompt injection attacks. This flaw enables attackers to execute arbitrary Python code, substantially compromising the security of the application. Instead of delivering the intended natural language processing results, the vulnerability can lead to unauthorized remote code execution, putting sensitive data and systems at risk. Developers using PandasAI should adopt immediate security measures to mitigate this risk.

References