Who else low-key wants Google Bard to get better than GPT-4?
I like Bard. It had a bad first impression but after Google threw in more PaLM magic to it, it has gotten much better. There are still times that GPT-4 produces more correct results. But I low-key wish Bard could get better than GPT-4. To some extent, Bard "feels" like a cleaner approach to creating safe LLMs whereas GPT-4 feels like a "hack".
Do you agree on these things?
Eventually I hope the compute required to run these systems become local, but yes I do find Bard's "personality" to be a little more controlled and cleaner than chatGPTs. However, chatGPT 3.5 usually gives better responses (e.g. specific book recommendations I ask it).
I agree in general, GPT 4 does feel like a hack compared to Bard. But last I tried (3 days ago) its answers were a bit worse than GPT 3.5 , at least for the specific questions I asked it.
Why do you feel that Bard is clean and GPT-4 is a hack?