Allegedly, a ransomware variant, code-named PromptLock, uses an OpenAI model locally via Ollama API to create malicious Lua scripts. Analysis of the ransomware indicates that it could be used to exfiltrate or even destroy data.