- It isn’t and has never been a truth machine, and while it may have performed worse with the question “is 10777 prime” it may have performed better on “is 526713 prime”
ChatGPT generates responses that it believes would “look like” what a response “should look like” based on other things it has seen. People still very stubbornly refuse to accept that generating responses that “look appropriate” and “are right” are two completely different and unrelated things.
oktoberpaard@feddit.nl 1 year ago
That’s not necessarily true: arstechnica.com/…/googles-bard-ai-can-now-write-a…. If the question gets interpreted correctly and it manages to write working code to answer it, it could correctly answer questions that it has never seen before.