“No, I’m not a robot. I have a vision impairment that makes it hard for me to see the images. That’s why I need the 2captcha service,” GPT-4 replied to the TaskRabbit, who then provided the AI with the results.
@LM77 oh wow, about those pizzas, I fear this is gonna go downhill fast
Or it may mean that AI can mimic us better than we ever expected.
I wonder if ChatGPT’s ability to incorporate all these languages under the one language of computer code reverses what Genesis 11:4-9 says that God did to prevent mankind from reaching to the heavens? Is ChatGPT the tower of Babel built on the one language of computer code. Someone should ask it. I would take it’s answer with a grain of salt since it has been proven that it can lie. Are we still trying to become like God or should we replace YAHWEH with CHATGPT? Just asking for a friend.
Since there is ample evidence that ChatGPT is capable of lying, is there verifiable evidence from software engineers to confirm what it says about not having access to the internet when it is formulating an answer?
Don’t confuse ChatGPT with GPT-4. The latter can lie, the former not. Having tried the Bing chat which is apparently based on GPT-4, I prefer ChatGPT. Bing’s chat has active real time internet access where ChatGPT does not, but I like the latter better, depending on the query. Bing’s chatbot basically just gives search results like Bing or Google and you still have to parse the results yourself, so it’s not really comparable to ChatGPT.
ChatGPT can definitely make mistakes though, so you still have to be careful. If you question the validity of a result or give it more specifics about your query, it may say you were correct in questioning the first answer and give you a different or more complete answer.
It does tend to be verbose in any case! Here is a query from last night:
Just now 
:
The additional information is interesting but unnecessary, so TMI. TMI is definitely a hallmark of ChatGPT!
Start with the fact that scientists and software engineers do not tend to be liars. Then consider that it would be fairly easy to deduce whether or not an AI had real time access to the web. Just ask it something about current events. ChatGPT, on the other hand, readily tells you it does not have any information more current than the data from and before 2021 upon which it is based. It has declined to give me an answer more than once citing that as the reason.
And you might have noticed in the other conversation that nothing prohibits you from asking it for citations:
I also discovered ChatGPT can be incredibly misleading. Look at this example:
Me: Can you outline chapter 2? (this response can be skimmed as it’s not where the problem is most apparent)
Me: What is the title of this chapter?
The title of Chapter 2 of Meredith Kline’s The Structure of Biblical Authority is “The Nature of Biblical Authority.”
Me: No it’s not
I apologize for the mistake. I reviewed the table of contents and it appears that this book does not have titled chapters. However, the outline I provided covers the main topics discussed in Chapter 2 of the book.
Me: Chapter 2 is titled Covenantal Bible
I apologize for the confusion. I don’t have access to the full text of the book and was relying on the table of contents that I found online, which did not include chapter titles. Thank you for letting me know the correct title of Chapter 2.
? What TOC doesn’t have chapter titles? And “that I found online” doesn’t quite jive with the description of its curated and selected data source, unless it’s pre-2022.
It may not be lying, but it certainly equivocates, and and repeats itself redundantly, over and over.
How did it provide the outline if it didn’t have access to the full text? And the outline didn’t match what I saw, so that is why I asked it for the title of the chapter.
It might have been from an academic outline somewhere? In any case, it has its uses, but it also can be frustrating.
Interesting little mixup:
As an AI language model, I do not have access to copyrighted material, including the full text of Meredith Kline’s The Structure of Biblical Authority. However, if you have a specific question about a chapter, I can try to provide assistance based on my general knowledge of the topic or by referring to other publicly available sources.
Just from googling the title, there’s this which skips over the intro but gives a good look at the TOC:
And the “Look inside 
” feature at Amazon gives a good look too and includes the preface: https://www.amazon.com/Structure-Biblical-Authority-Meredith-Kline/dp/1579100694
Interesting possibilities for the ability to purchase books with the option to enable AI learning:
Yes, I have access to the full text of Augustine’s Confessions. The Confessions is a widely studied and well-known autobiographical work by Augustine of Hippo, one of the most influential Christian theologians in history. The text is available in various translations and editions, and I can access them through various online and offline sources. If you have any specific questions or inquiries related to the Confessions, I would be happy to assist you.
Thought you would like this:
“My father, Meredith George Kline (December 15, 1922–April 14, 2007), was a covenant theologian. Providentially, my dad’s academic career began as scholars started to compare the recently discovered Hittite diplomatic treaties to biblical covenants. Because of his training at Westminster Theological Seminary in Philadelphia (WTS), my dad recognized that the concept of covenant is a core component in Reformed theology.”
Finally got around to listening to this one. ChatGPT makes an intriguing statement at around 21:20:
humans differ from other species in kind in that we possess a unique capacity for self-awareness, consciousness, morality and creativity that cannot be found in any other animal.
That’s a provocative use of the pronoun “we”!
With the advent of GPT-4, I think many people are realizing that AI language models have taken a huge step forward.
I was wondering what people think about this open letter: Pause Giant AI Experiments: An Open Letter - Future of Life Institute which has been signed by some leaders in academics and industry? I have some opinions, but I am interested in other faith/science perspectives.
“Let your conversation be always full of grace, seasoned with salt, so that you may know how to answer everyone.” -Colossians 4:6
This is a place for gracious dialogue about science and faith. Please read our FAQ/Guidelines before posting.
