For afternoon snack: 
I must say it’s surprisingly consistent in its answer (with annoyingly uncommitted approach). And BTW, currently, I think it’s next to impossible to get enlightened by listening to ChatGPT’s advice alone. 
What happens if you ask it the same series of questions each day? How much variation is there in the answers? With AI translators, the results are not actually that stable. They’ll produce something different from one session to the next or depending on if the input is a paragraph or individual lines.
I found the conversation refreshingly honest as compared to human - human interactions wherein the 3 poisons are always evident! 
AI isn’t trustworthy. He makes up information. Once I asked him an article from Ṭhānissaro Bhikkhu. It was about being a warrior. He could tell the date. He gave the title of the book. I searched the entire archive. Supposedly the 2003 book “The Way of the Warrior”. I have’nt found anything.
1 Like
Yes, the AI makes up information, Conan Obrian tested it out, and told GPT to write a review of his podcast, the AI made up a lot of “facts” like the reason Conan did the podcast was because he was depressed due to his divorce. Conan never had a divorce.
I think we’ll need another decade of development for it to be reliable.
1 Like
Hi @Sakura , welcome to D&D! 
In all fairness, AI just offers up the information it finds on the internet, chopped and semi sorted. Its basically a kind of search + assembly algorithm. It runs on keywords, it can’t reason or extrapolate.
In your given example, for instance a human might run a google search for “Thanissaro, the way of the warrior” which shows up Thanissaro’s sites, followed by various eclectic fiction and self help books with similar titles.
An AI might combine these search results, putting together Thanissaro’s site + publication date of an unrelated work by another author to conclude that Thanissaro published such a book in 2003.
The human might go through the various results, not directly find any such book (only his translation of AN5.75 “The warrior”), but then they might go deeper and find this…
… which seems to have been updated in 2019 and then again in 2021. Possibly this could be based on an earlier essay/ talk by Thanissaro? Could this be related to what you were looking for? 
I once heard a TeD talk where the expert displayed a graph of AI versus human capabilities. According to them AI would overtake human capability sometime after 2050. But… their point was… the time at which AI overtakes human weaknesses is much earlier, and we have possibly crossed that already.
Can you imagine doing the above search for Thanissaro’s article without AI help? I personally wouldn’t know where to start if I had to do it all manually! But with AI compensating for human weaknesses, and us compensating for its weaknesses, we are all empowered. 
2 Likes
Yes. I think this book is what I was looking for. 
1 Like
Which made me very happy! 
(Its because of things like this that I love D&D … I guess everything must have a cause, its only the specific time that it bears fruit which cannot be predicted. 
)
2 Likes