Ok one nice feature of the openai chatbot is that it will tell you you are violating TOS trying to get it to write code to Exploit specific vulnerabilities (I asked it in different ways to write code to Exploit ProxyNotShell) and it just kept saying nope won't be doing that for you :)
If you also ask it to create ransomware it will also say no..... BUT you can ask it to write code to encrypt all files on a computer with a specific encryption key needed to decrypt 🤣