Re:A New Chat Bot Is a ‘Code Red’ for Google’s Search Business
by nprateem on Saturday December 24, 2022 @03:50AM
(#34115150)
I think solving this inaccuracy will be openai's biggest challenge. The web is safe for now. Once they do fix it it's basically game over.
Re:A New Chat Bot Is a ‘Code Red’ for Google’s Search Business
by ilaksh on Saturday December 24, 2022 @08:42AM
(#34116700)
You can turn the temperature down on the OpenAI API and it is much more accurate. They also have a code-specific model which I don't think ChatGPT is using.
Re:A New Chat Bot Is a ‘Code Red’ for Google’s Search Business
by moritonal on Saturday December 24, 2022 @05:21AM
(#34115571)
But, you have to admit that if ChatGPT had a process whereby it'd internally list the sites most likely relevant to your question (essentially Google the keywords) then included those sites as context it'd likely ace your question?
Basically, if Google owned ChatGPT the result would be very different.
Re:A New Chat Bot Is a ‘Code Red’ for Google’s Search Business
by spaceman_2020 on Saturday December 24, 2022 @05:55AM
(#34115743)
Whenever chatGPT gives me a function and I try it out and it doesn’t work, I just tell it “hey this isn’t working” and it will correct itself
This is exactly the scenario I encountered. I have a teammate who is using ChatGPT to ask questions as opposed to searching documentation and it gave incorrect info all around on all the following:
1. Where the property would be found in the event schema 2. What structure the value would be 3. The fact that it's not included at all for AWS AppSync specifically in question, even if the previous two had been answered correctly regarding API Gateway event schema.
All of this was available in the first doc returned in the search results of both API Gateway and AppSync respectively.
The fact it convincingly tells you otherwise causes additional friction since they're pointing at their screen telling me I'm wrong cause look at the response from ChatGPT.