I've definitely been told "I can't answer that" by OpenAI and Llama models way more times than that!
Deepseek > Chat-GPT = Llama >>> Claude.
Deepseek seems like a neutral tool. Claude is very, very preachy. I'm happy that there doesn't appear to be any reason to ever use it again.
Whether this is a good idea is up for a debate, but it seems to have worked well in my case (I haven't had my app ddos'd, so I don't know for sure.)