3 ms·
I've definitely been told "I can't answer that" by OpenAI and Llama models way more times than that!
by bbqfog 2y ago
I've definitely been told "I can't answer that" by OpenAI and Llama models way more times than that!
- ggregoire 2y agoI've been trying this week to summarize transcripts from Fox News with llama3.1 and half the time it tells me it can't because this is too sensitive…
- A_D_E_P_T 2y agoClaude is the worst. It can barely even tell jokes. In terms of "openness" and "willingness to respond" I rank 'em: 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.
- all2 2y agoClaude works well for basic code boilerplate generation. I got firewall rules and nginx config for a basic app deployment just last night. 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.)
- blackeyeblitzar 2y agoI agree that Claude can be very preachy and frustrating. But I disagree that DeepSeek is the most neutral. I think it is actually the least neutral because the censorship is by design and forced by the government. Claude basically has clumsy safety features that are not sophisticated to stay out of the way but without malicious intent, unlike DeepSeek.