A community to discuss about ChatGPT and AI. Not affiliated with OpenAI.
Rules:
NSFW must be marked as such
No porn
Don't be toxic
jailbreaks are allowed for now at least
No trashposts.
Keep the memes to a minimum
To me this software carries a sense of the absurd.
I think AI bots are about trying to "think less", or at least do less work in research & analysis on a question (hallucinations overlooked for now). Having to analyze & decide between many different AI answers doesn't accomplish that.
Although I guess being able to correlate answers between bots might allow you to catch a hallucinating one.
Yes! That's one of the main uses - When I make a query, I find that none of the bots tend to give "the most" comprehensive answer- typically I find that Claude and ChatGPT4 give the best answers (often Claude is superior) though one of the other will miss out one or two bits of information, and sometimes (often?) Bing suprises and gives the best reply.
As we are seeing the infancy of these services, I find this program ideal for testing the bots to eventually make a decision of which ones to use as the goto.
I am also trying to determine whether to continue with a ChatGPT plus subscription, and after making a lot of tests (including ranking results from 6 bots), I found that Claude had the highest score, then CHatGPT4, then Bing, then Chatgpt/bard, so when my subscription for the month comes up I will liely let it lapse and stick with Claude/bing (+/- chatgpt-3.5 and bard).
If you are doing anything where results are more crucial (anything from more serious research, queries for work or asking the bots to compare the top 5 printers and rank according to price per page, then having multiple anwers I find are VERY useful, as they all hallucinate, as you say).
Thank you for sharing this!
Sure! Chatgpt4 has been altered a bit so they're trying to get around it, but it's my goto softwarre for AI chats.
