This will supplement two of ChatGPT's larger deficiences: math and the veracity of its output. It'll be interesting to see how they price this integration, and how long OpenAI will maintain the $20 pricing for ChatGPT+.
ChatGPT can now call Wolfram Alpha
81–90 of 329 posts
Re: ChatGPT can now call Wolfram Alpha
#82Confirmation that Google is dead.
For the first time in a decade they have a viable competitor. They must figure it out and I'm confident they will. They understand they're quite literally are facing extinction. I'm sure they're in focused mode since the original ChatGPT announcement, this should get them into proper creative panic mode.
Microsoft has a very good leader at the helm, Alphabet does not. You can already see the difference that makes from the last few years. Alphabet has no soul culturally, whatever it once had is long gone.
Microsoft is overwhelmingly a software company. What is Alphabet? Microsoft still knows what it is.
Re: ChatGPT can now call Wolfram Alpha
#83OpenAI is like Google before 2003, dazzling the world with amazing products backed by astonishing technologies that went beyond wild dreams of millions of people. And most importantly, OpenAI kept releasing new features at blazing speed with a few hundred people, just like early days of Google and Microsoft.
This reminds me of stark contrast of Google's GSuite and Microsoft Office. Microsoft released new office every couple of years, introducing huge improvements and packing tons of features[1]. In contrast, GSuite appeared to be, hmm, quite quiet.
[1] You may argue that no one needs that many features, but I subscribe to Joel Spolsky's idea: https://www.joelonsoftware.com/2001/03/23/strategy-letter-iv...
Re: ChatGPT can now call Wolfram Alpha
#84AFAIK Wolfram was offered $1B acquisition by MS back when it initially released in order to incorporate it into Bing. Obviously he declined. I wonder what they paying to use it now.
Could be just zero, because they can consider that given the very very large popularity of OpenAI and the niche popularity of WolframAlpha, that OpenAI is doing free advertisement to WA.
Re: ChatGPT can now call Wolfram Alpha
#85Does anyone know how ChatGPT is deciding is should use Wolfram? I’m curious for two reasons 1) previously it didn’t seem to ‘know’ when it was wrong, so it must have some way to detect this now. 2) when there are more (potentially competing competing?) plugins or will be interesting to see how it decides which is the one to use.
With a fairly naive ReAct implementation with access to, e.g., Wikipedia, GPT-3.5-turbo will decide it should confirm things that it already thinks it knows (one test I remember it doing this on was “What is the largest land animal?” where the “Thought” portion of its response was approximately [wasn’t recording the runs] “I believe its the elephant, but I should confirm”.) So, it doesn’t have to recognize it is wrong to decide that it should use available resources to confirm.
Re: ChatGPT can now call Wolfram Alpha
#86RIP homework
Homework: No Proven Benefits: https://www.edutopia.org/no-proven-benefits
Re: ChatGPT can now call Wolfram Alpha
#87Very cool. Effectively, ChatGPT is using tools now. I wonder how much longer before it evolves opposable thumbs :)
Probably Elon Musk is going to connect arms and legs to a LLM like ChatGPT, add a fart sound effect, and call it Tesla Bot.
Re: ChatGPT can now call Wolfram Alpha
#88Does anyone know how ChatGPT is deciding is should use Wolfram? I’m curious for two reasons 1) previously it didn’t seem to ‘know’ when it was wrong, so it must have some way to detect this now. 2) when there are more (potentially competing competing?) plugins or will be interesting to see how it decides which is the one to use.
My assumption: It gets input of real time data from a live source (Wolfram Plugin). It also has lots of data and sources gathered way back when its ?original? data was crawled. And then it evaluates which one is more "precise".
Bing AI also has options: Creative, Balanced, and Precise. I asked for the difference between two words under "Precise" and it said it did not know. I changed it to "Balanced" and it gave an answer that was pretty much text book definitions of both words.
I asked Bing AI to pretend to be an intellectual on AI concepts. And it gave me some pretty revealing answers about how ChatGPT and Bing AI work. It is suffice to say I know longer think of these AIs as magical anymore. It is just better able to understand what people are _asking_, and then know to give responses that are a direct answer to people's questions.
Re: ChatGPT can now call Wolfram Alpha
#89Earlier quoted context omitted.
> They missed the first-mover advantage with Cloud They also missed the first-mover advantage with web search. And webmail. And online video hosting. And online ads. (In fact, very few of the dominant players anywhere in tech had first-mover advantage in that field.)
They bought their way into several of those, something that is much more difficult to to do these days.
Microsoft have effectively bought their way into OpenAI. The fact Google have done LaMDA on their own gives them a great starting point to compete.
OpenAI could just be another Netscape.
Re: ChatGPT can now call Wolfram Alpha
#90Does anyone know how ChatGPT is deciding is should use Wolfram? I’m curious for two reasons 1) previously it didn’t seem to ‘know’ when it was wrong, so it must have some way to detect this now. 2) when there are more (potentially competing competing?) plugins or will be interesting to see how it decides which is the one to use.
Ironically I had implemented this myself with the GPT 3.5 API:
Wolfram is a good match because it already supports fuzzy searches, has a well formed API, and the API responses provide information in a separate objects sorted by importance.
That last bit is important because if you just open a random website, there's almost no chance it fits in the context window, and if you just truncate the contents there's no guarantee you capture important info. I ended up messing around with using non-LLM NLP to summarize webpages, but it's still a pain.
-
As far as knowing, ChatGPT doesn't need to know it's wrong, it just needs to know what it tends to be wrong at.
For example, I told my instance: "Any time you feel like the specificity of your answer would be improved, you may use !q ", "Any time a question would benefit from more recent information you may use !q " and "As a Large Language Model your math cannot be trusted, use !q ".
That was enough to get it using Wolfram for any math or factual searches