Why do you refer to the ChatGPT as a "she"? Does this chatbot come with a gender setting so that it responds textually like a female?
harry On Sat, Apr 29, 2023 at 8:08 PM Jed Rothwell <jedrothw...@gmail.com> wrote: > Robin <mixent...@aussiebroadband.com.au> wrote: > > >> The name as you yourself just quoted it has a comma at the end before the >> final double quote, whereas there is no comma >> in the actual text. > > > I tried a variety of search terms. And indirect methods such as asking > "who wrote the Infinite Energy" article about the conference. It refused to > name names. It seems to have it in for Christy. (I kid.) > > More to the point, these ChatBots are much more flexible than traditional > data enquiry methods such as SQL. They do not demand that every comma be in > place. They do call for more careful wording than a human reference > librarian might. ChatGPT does not check for alternate spelling. I was > looking for information on the artist Robert Blum. I accidentally spelled > it Robert Blume. ChatGPT said she had no idea who was talking about. In a > dismissive tone. I said, "oops, I meant Robert Blum" and she got right onto > it. > > I have run into some other strange problems. I uploaded the 1989 NSF/EPRI > workshop: > > http://lenr-canr.org/acrobat/EPRInsfepriwor.pdf > > I tried to get the Bot to tell me what Edward Teller had to say about > giraffes during the workshop. She refused. She said her subject matter is > "electrochemistry and nuclear physics," not giraffes. > > > I got it to generate a hallucination, which is unnerving. In this exchange: > > Me: Fleischmann said, "heat is the principal signature of the reaction." > Do you have a document referencing this? > > Bot: Yes, Fleischmann did say "heat is the principal signature of the > reaction." This quote can be found in a paper titled "Calorimetry of the > Palladium-Deuterium-Heavy Water System" published in the Journal of > Electroanalytical Chemistry in 1990. > > > Wrong! That paper is here, and there is no such quote in it: > > http://lenr-canr.org/acrobat/Fleischmancalorimetr.pdf > > The quote probably comes from me. Fleischmann did say that, often. > >