DeadJoe
Elite Member
- May 18, 2014
- 2,235
- 2,992
It is impossible for them to test this for every single topic. They can improve accuracy but I doubt it will ever be 100%, people will get incorrect answers. Let's see what they do next.
LOL @ Garbage Google AI
- Thread starter DeadJoe
- Start date
It is impossible for them to test this for every single topic. They can improve accuracy but I doubt it will ever be 100%, people will get incorrect answers. Let's see what they do next.
coldoutcome
Newbie
- Jun 6, 2024
- 17
- 7
And OpenAI just announced partnership with Reddit lmao. Why would they ever admit to this
- Mar 22, 2013
- 6,272
- 10,985
I have tried several ai models, also Google Gemini Pro 1.5 and I have to say it is bad compared to Claude, ChatGPT or Grok.
Its quite interesting, because Google probably has the most collection of data, the most real time access to any Android device, Google Chrome browser, and so on,.
Its quite interesting, because Google probably has the most collection of data, the most real time access to any Android device, Google Chrome browser, and so on,.
coldoutcome
Newbie
- Jun 6, 2024
- 17
- 7
at this point it seems like google just falls over and over again. what have they done since 2018 or so? been a while since they came out with something new and successful.
- Oct 28, 2020
- 11,663
- 14,635
just wanted to say that the quote in your signature is so true. I wish everyone saw and understood it...
AshOasis
Regular Member
- Jan 8, 2024
- 297
- 572
Of course not, but even according to their own guidelines, website owners have to be especially mindful of any topic regarding health and medicine. And then they come out and suggest you to eat glue, lol.
It sure is, of course, it's not my quote but by Carl Jung.
- Jun 16, 2022
- 2,832
- 852
Now small companies are able to defeat google products in certain categories.. That is not a good sign for them.
Stewartklose
Failed to resolve dispute
- Jun 23, 2014
- 149
- 50
there is still too much work in AI field
SuperProxyShop
Junior Member
- Nov 6, 2020
- 121
- 23
The funniest thing is that Google invests so much effort in evaluating DA, DR, and such, but still even one random page about scissor running that was indexed some time ago can disrupt all experience of using their AI
Seems as if they didn't even test it before launch on various use cases
To Google's credit, however, almost all those queries are some trash talk, so it isn't really about bad resluts. It's about inability to distinguish funny query from a serious one
Seems as if they didn't even test it before launch on various use cases
To Google's credit, however, almost all those queries are some trash talk, so it isn't really about bad resluts. It's about inability to distinguish funny query from a serious one
When_tomorrow_comes
Junior Member
- Mar 29, 2024
- 142
- 52
Wow, this AI went way further down the rabbit hole than was intended))
Sughwate
Banned - Multiple accounts to game the forum.
- Feb 27, 2023
- 380
- 29
Why? I don't understand. Are they usually incorrect?
- May 27, 2016
- 11,083
- 2,516
If you actually knew how chatgpt works you would know its answers are based on a predictive model, always predicting what words should come next. It's the same how we learn to speak as toddlers. The more some words are used the more it puts them into sentences sometimes leading to mistakes such as these.
avanaart22
BANNED
- Nov 4, 2023
- 24
- 44
one of the worst use cases of AI
DeadJoe
Elite Member
- May 18, 2014
- 2,235
- 2,992
I am actually quite familiar with predictive models (my 9 to 5 is analytics). That doesn't change anything, the statements are just factually incorrect. Displaying incorrect information on such a mainstream medium isn't smart, regardless of the underlying technology.