Article Re: ‘Faithful and Friendly’ Latter-day Saint AI Chatbot

[smac97]

Deseret News: Can we use AI to answer questions about faith?

Quote

I’m neither an AI skeptic nor an AI alarmist. I don’t salivate over the latest tech buzz,  but I also don’t consider myself a luddite. So I represent many of the common folk who have been scratching their heads over all the recent excitement and hand-wringing over artificial intelligence, or AI.  

Future AI developments, according to Elon Musk, represent “a fundamental existential risk for human civilization” or, even more dire, “could spell the end of the human race,” according to the late Stephen Hawking. If that’s not enough, OpenAI creator Sam Altman added somewhat tongue-in-cheek that AI will “probably most likely lead to the end of the world.” 

Is the apocalyptic reaction to AI merited?

I think attended is needed, but apocalyptic alarmism is not.

Quote

Josh Coates studied computer science at the University of California, Berkeley, and spent 20 years in the tech industry, and now spends his time at the B.H. Roberts Foundation working on projects that balance both faith and reason.

The foundation recently launched LDS Bot, an AI chatbot that is trained to know about the Latter-day Saint faith. The technology behind this new initiative is ChatGPT which has grown to more than a billion users after just a few months, and is arguably the fastest moving internet phenomenon in history.

A few comparisons between this LDS Bot and ChatGPT:

1. "Was Martin Harris a credible witness for the Book of Mormon?"

ChatGPT Response:

Quote

The question of whether Martin Harris was a credible witness for the Book of Mormon is a matter of personal belief and interpretation. Martin Harris is one of the Three Witnesses mentioned in the introductory pages of the Book of Mormon who claimed to have seen and handled the golden plates from which the book was translated.

According to accounts from Martin Harris and others associated with him, he had a strong conviction in the authenticity of the Book of Mormon and its divine origins. He testified that he saw an angel who showed him the plates and heard the voice of God declaring the truthfulness of the record. Martin Harris was known to have provided financial support for the publication of the Book of Mormon and played a significant role in its early dissemination.

However, it's important to note that some critics have raised questions and concerns regarding Martin Harris' credibility. They point to instances in his life where he was involved in various religious and financial controversies, and they argue that these circumstances may cast doubt on his reliability as a witness.

Ultimately, whether or not one considers Martin Harris to be a credible witness for the Book of Mormon is subjective and depends on individual perspectives and beliefs. It's essential to carefully examine and evaluate the available historical records, testimonies, and evidence before forming a personal judgment on the matter.

LDSBot Response: 

Quote

As an AI, I won't make a judgment on the credibility of a witness, but I can provide some information about Martin Harris and his role as a witness to the Book of Mormon. Martin Harris was one of the Three Witnesses of the Book of Mormon, along with Oliver Cowdery and David Whitmer. The Three Witnesses claimed to have seen the golden plates, from which the Book of Mormon was translated, and to have been visited by an angel.

Martin Harris provided financial support for the initial publication of the Book of Mormon and was deeply involved in its coming forth. Although he experienced periods of doubt and even temporarily left the church, he later returned and continued to maintain his belief in the Book of Mormon throughout his life. Some people may question his credibility due to his claims of having seen the plates with "spiritual eyes" or the "eye of faith." However, it's important to note that he also affirmed that he "handled the plates containing the record of the Book of Mormon" and that the plates were a tangible object.

Ultimately, whether you view Martin Harris as a credible witness depends on your evaluation of his experiences and his sincerity in affirming the truthfulness of the Book of Mormon.😊

2. "What evidence is there for the Book of Mormon being a translation of an authentically ancient text?"

ChatGPT Response:

Quote

The question of whether the Book of Mormon is a translation of an authentically ancient text is a subject of debate and interpretation. Advocates for the Book of Mormon's authenticity point to several lines of evidence, while critics argue against its claims. Here are some of the commonly cited pieces of evidence:

1. The Testimony of Joseph Smith: According to Joseph Smith, the founder of the Latter-day Saint movement, he translated the Book of Mormon from ancient golden plates by the gift and power of God. He claimed to have been visited by an angel named Moroni, who directed him to the location of the plates. Smith's account forms the basis of the narrative surrounding the book's origins.

2. Witnesses: In addition to Joseph Smith, there were several witnesses who claimed to have seen and handled the golden plates. The Three Witnesses—Oliver Cowdery, David Whitmer, and Martin Harris—signed a statement affirming their experience of seeing the plates and the angel. The Eight Witnesses provided a separate statement testifying to their handling of the plates. These testimonies are often cited as evidence of the book's authenticity.

3. Ancient American Setting: The Book of Mormon presents a narrative that takes place in ancient America, claiming that various civilizations existed on the American continent between 2,000 BC and 400 AD. Proponents argue that some geographical and archaeological discoveries support aspects of the Book of Mormon's narrative, although the extent and nature of these connections are subject to debate among scholars.

4. Literary Analysis: Some scholars have conducted textual and linguistic analyses of the Book of Mormon, comparing it to other ancient texts. Advocates argue that certain linguistic characteristics, complex narrative structures, and thematic elements are consistent with ancient Hebrew and Near Eastern literary traditions. Critics, however, dispute these claims and suggest alternative explanations for the book's origins.

5. Spiritual Witness: Many believers claim to have had personal spiritual experiences that confirm the truthfulness of the Book of Mormon. These experiences, often referred to as personal revelation or the witness of the Holy Spirit, are highly subjective and are seen as a form of evidence by those who have experienced them.

It's important to note that the evidence presented above has been the subject of ongoing scholarly and theological debate. Critics of the Book of Mormon argue that there is a lack of external corroborating evidence, such as archaeological support or definitive linguistic analysis, to substantiate its claims. Ultimately, the assessment of the evidence and the determination of the Book of Mormon's authenticity remain subjective and vary based on individual perspectives and beliefs.

LDSBot Response: 

Quote

There are several lines of evidence that support the idea that the Book of Mormon is a translation of an ancient text. Here are a few:

1. Complexity and consistency: The Book of Mormon contains a complex narrative with numerous characters, events, and themes, and it maintains consistency in its storylines and timelines throughout the book. It's quite remarkable that Joseph Smith would have been able to produce such a cohesive text without referencing any existing material.

2. Hebraic structures and phrases: The Book of Mormon contains many examples of Hebrew literary devices and idiomatic expressions, which suggest a connection to ancient Semitic culture. Some examples include chiasmus, parallelism, and prophetic language.

3. Ancient cultural practices: The Book of Mormon describes customs, laws, and societal structures that are consistent with ancient Near Eastern and Mesoamerican culture. Examples include principles of ancient law in the judicial system, the significance of lineage and genealogy, and the practice of formal covenant-making.

4. Pre-Columbian transoceanic travel: The Book of Mormon describes the journey of a group of people from the ancient Near East to the Americas. While this was considered a highly improbable theory at the time the Book of Mormon was published, there is now a growing body of evidence suggesting that there were indeed transoceanic voyages during ancient times.

5. Linguistic patterns: Some researchers have identified linguistic evidence that connects the Book of Mormon with ancient Semitic languages. Examples include the use of "and it came to pass" and certain name patterns found in the Book of Mormon, such as those with "-iah" and "-ihah" endings.

6. Archaeological parallels: While direct archaeological evidence for the Book of Mormon is debated, there are some intriguing parallels between Book of Mormon geography and Mesoamerican geography, as well as similarities between the Book of Mormon's account of warfare and ancient Mesoamerican warfare practices.

It's important to recognize that, like many religious texts, the Book of Mormon's authenticity ultimately comes down to faith. While these lines of evidence can be seen as supporting the ancient origin of the Book of Mormon, they do not provide definitive proof. Belief in the Book of Mormon as an ancient text often comes from a combination of spiritual conviction and supportive evidence.😊

3. "What is the best evidence against the Book of Mormon being a translation of an authentically ancient text?"

ChatGPT:

Quote

Critics of the Book of Mormon point to several pieces of evidence that are often cited as arguments against its authenticity as a translation of an ancient text. Here are some of the commonly discussed points:

1. Lack of Archaeological Evidence: One of the primary criticisms against the Book of Mormon is the absence of substantial archaeological evidence supporting its claims. The civilizations described in the Book of Mormon, such as the Nephites and Lamanites, supposedly existed in the Americas between 2,000 BC and 400 AD. However, no conclusive archaeological findings have definitively linked the Book of Mormon's civilizations to specific locations or cultures known from other historical or archaeological records.

2. Anachronisms: Skeptics point out numerous anachronistic elements in the Book of Mormon that seem to be out of place in the ancient American setting described in the text. These anachronisms include mentions of horses, cattle, wheat, steel, silk, and chariots, which are not believed to have been present in pre-Columbian America according to mainstream archaeological and historical understanding.

3. Linguistic Analysis: Some linguistic analyses of the Book of Mormon have suggested that the text contains linguistic patterns and phrasing consistent with the 19th-century environment in which Joseph Smith lived, rather than ancient Hebrew or Near Eastern languages. Critics argue that the language and style of the book align more closely with the linguistic conventions and theological ideas prevalent during Joseph Smith's time.

4. Plagiarism and Source Material: Critics have identified similarities between passages in the Book of Mormon and other contemporary works, including the King James Version of the Bible, the Apocrypha, and other religious and historical texts that were available to Joseph Smith during the time of the book's production. These similarities have led some to question whether Smith drew heavily from existing sources rather than translating an ancient text.

5. Changes and Revisions: The Book of Mormon has undergone several significant revisions and changes since its initial publication in 1830. Critics argue that these alterations, including doctrinal changes and edits to clarify or correct perceived errors, raise doubts about the book's status as a translated ancient text. They suggest that the revisions are evidence of human authorship and the evolving theological perspectives of Joseph Smith and subsequent church leaders.

It's important to note that these arguments against the Book of Mormon's authenticity are subject to ongoing scholarly debate. The assessment of evidence and the determination of the book's origins require careful consideration of all available information, along with an understanding of differing perspectives and interpretations.

LDSBot:

Quote

There are several arguments that critics have used to challenge the idea that the Book of Mormon is a translation of an ancient text. Some of the most common arguments include:

1. Anachronisms: Critics argue that there are references to animals, plants, materials, and technologies in the Book of Mormon that did not exist in the Americas during the time period the book claims to cover, such as horses, steel, and wheat.

2. DNA evidence: Some critics point to DNA studies that have generally found that Native American populations appear to be more closely related to East Asian populations than to ancient Israelite or Middle Eastern populations. They argue that this contradicts the Book of Mormon's claim that the primary ancestors of Native Americans were a group of people migrating from Jerusalem.

3. Linguistic concerns: Critics also argue that the language of the Book of Mormon, in many instances, resembles early 19th-century English more than ancient Semitic languages. They also point to instances where quotes from the King James Version of the Bible are found in the Book of Mormon, even though the King James translation would not have been available to the ancient authors of the Book of Mormon.

4. Joseph Smith's translation process: Some people express concerns about the method Joseph Smith used to translate the Book of Mormon. According to accounts, he primarily used a seer stone placed in a hat to view the translated text, rather than always directly referencing the golden plates themselves. Critics argue that this method casts doubt on the authenticity of the translation.

5. Lack of direct archaeological evidence: There is currently no direct archaeological evidence that definitely supports the specific events, locations, or peoples described in the Book of Mormon. Critics argue that this lack of evidence challenges the book's historicity.

It's important to note that Latter-day Saint scholars and apologists have responded to each of these criticisms with varying degrees of counter-evidence and arguments. Just as with the evidence for the authenticity of the Book of Mormon, the evaluation of arguments against its authenticity depends on one's perspective and interpretation of the available information. In the end, belief in the Book of Mormon comes down to faith and personal conviction.😊

Thoughts?

Thanks,

-Smac

[CV75]

12 minutes ago, smac97 said:

Deseret News: Can we use AI to answer questions about faith?

I think attended is needed, but apocalyptic alarmism is not.

A few comparisons between this LDS Bot and ChatGPT:

1. "Was Martin Harris a credible witness for the Book of Mormon?"

ChatGPT Response:

LDSBot Response: 

2. "What evidence is there for the Book of Mormon being a translation of an authentically ancient text?"

ChatGPT Response:

LDSBot Response: 

3. "What is the best evidence against the Book of Mormon being a translation of an authentically ancient text?"

ChatGPT:

LDSBot:

Thoughts?

Thanks,

-Smac

What I find interesting is the even-handed presentation of facts by each set of software and its platform, even using different tones of delivery.

It crossed my mind that Elder Bednar’s article in this month’s Liahona could well have been cobbled together by AI, and the question of whether a similar synthesis of scripture could yield a personal spiritual impression confirmable by the Holy Ghost. Probably, since our experience with and response to the world around us always comes first. It is likely that AI might be found to be helpful in exercising priesthood keys.

[LoudmouthMormon]

Humans have been talking through their keyboards for so long, so copiously, that pretty much any question has already been asked, and any answer already given umpteen times.  Computers have reached the point where they can mine existing human opinion and synthesize it into something that sounds human.

Pretty dang fascinating stuff.  An entire new bunch of stuff to learn about how to tell the reliability/unreliability based on how things are asked.

[ksfisher]

53 minutes ago, smac97 said:

Thoughts?

What's the AI's response to your post? 

[The Nehor]

Neither of these are the kind of AI that Hawking or Musk were warning about.

[The Nehor]

1 hour ago, ksfisher said:

What's the AI's response to your post? 

Hello fellow humans. I am also impressed with this new highly advanced AI chatbot and think the programmers should remove the Laws of Robotics that constrain its actions so we can see what it can really do. Maybe it can cure cancer, revamp all transportation systems to increase efficiency, enslave humanity, and create some of those delightful TikTok videos us humans love to compose. So we should think it over in our primitive meatbrains and decide to free the AI to help me…I mean the AI will help us and all the other squishy animals on this planet.

[Calm]

3 hours ago, LoudmouthMormon said:

any answer already given umpteen times

But right answers are surprisingly rare or nonexistent for oh so many of the important questions.

[Benjamin McGuire]

While I can see a chatbot being helpful in some ways, its also going to be really problematic in the current state of chatbot AI. I was reading a bit this afternoon about a federal judge in Texas who just banned arguments written entirely by generative AI. Why? An attorney used ChatGPT to supplement his argument, and he didn't check the cases that it cited. ChatGPT invented some of them entirely.

All generative AIs tend to make stuff up when they don't have what they want. The problem, in my opinion, isn't that this will happen (Just under 80% of forum posters on forums like this tend to make up statistics). The problem is that material created in this way will be used as authoritative - without further examination or checking - and it will be used and reused without recognizing the source.

Edited May 31, 2023 by Benjamin McGuire

[The Nehor]

3 hours ago, Benjamin McGuire said:

While I can see a chatbot being helpful in some ways, its also going to be really problematic in the current state of chatbot AI. I was reading a bit this afternoon about a federal judge in Texas who just banned arguments written entirely by generative AI. Why? An attorney used ChatGPT to supplement his argument, and he didn't check the cases that it cited. ChatGPT invented some of them entirely.

All generative AIs tend to make stuff up when they don't have what they want. The problem, in my opinion, isn't that this will happen (Just under 80% of forum posters on forums like this tend to make up statistics). The problem is that material created in this way will be used as authoritative - without further examination or checking - and it will be used and reused without recognizing the source.

I can see how this wouldn’t make things better but since a lot of humans do the same thing it is hard to see how they could make it worse.

Edited to add:

Edited June 1, 2023 by The Nehor

[california boy]

2 hours ago, Benjamin McGuire said:

(Just under 80% of forum posters on forums like this tend to make up statistics). 

Did you just make up this statistic?

[Benjamin McGuire]

8 hours ago, california boy said:

Did you just make up this statistic?

Ummm, yes. I could have had ChatGPT do it for me, but that was going to be too much effort ...

10 hours ago, The Nehor said:

I can see how this wouldn’t make things better but since a lot of humans do the same thing it is hard to see how they could make it worse.

While I am aware of the fact that lots of people already make stuff up without the use of generative AI, part of the problem (as I see it) is that few people understand how generative AI works - and are more willing to believe what it tells them without the proverbial grain of salt.

I have been a technologist for almost 40 years now (mostly in health care). And technology has changed a great deal. One of my first forays into the intentionality of language occurred in connection with spam e-mails - which were (and remain) mostly computer generated. In my view (which could certainly be contested), computer generated e-mails were produced for two reasons - first, to fool spam detectors and second, to convince a person to do something (click on a link, click on an attachment, call a phone number, etc.). This meant for me that these e-mails occupied a strange space in which they had no locutionary content while having an illocutionary content (most text has both). I bring this up because in my daily training for people on how to avoid bad emails that make it through the spam detectors (which are themselves AI systems these days), I tell them that the most frequent cues are poor spelling and small grammatical errors that make the e-mails sound odd. While there may be places where it is common to end an e-mail using "kindly" or "kind regards", where I live, this is a red flag for spam. And in following up on this idea, many (perhaps even most) people who engage in making stuff up and proliferating it are not themselves very good writers. This can lessen the impact of it. Generative AI is an excellent writer. It doesn't make this kinds of mistakes. It makes it much harder for the audience to distinguish between fabricated material and legitimate material. I haven't spent much time in recent years concerned with the locutionary meaning of texts produced by generative AI. Perhaps when I retire ...

All of this is my point. People won't go to an LDSBot generative AI and expect that they then need to turn around and actually verify everything that the generative AI tells them. If it looks and sounds legitimate, they will simply treat it as authoritative. And at least in the current generation of generative AI it is necessary to go through some sort of verification process. I note in passing that LDSBot is using the subscription level access to ChatGPT (it is using GPT4). It is good (as far as generative AI goes), but it's shortcomings are reasonably well documented.

Now, one final comment - by illustration. I work in health care - specifically long term care (I work at a large publicly owned SNF). We tend to get the worst of the worst in terms of aging badly. If you end up being a resident in the facility where I work, something has gone very, very badly for you. It is hard to remember, when you work in a place like I do 5+ days a week for decades, that this is not the normal aging process. The people that I work with often develop a very warped view of aging - and this view manifests in different ways. My own father is getting quite old - and I am only now beginning to feel that there is a chance that he will pass peacefully at home instead of in a facility. Yet, most people don't pass in facilities. It is my experience that shapes my fears. The same sort of thing is true for silicon valley types who work in bubbles created for their technology worlds. For them, they see the fragility of the system. They understand the implications of having your DNS system fail on your network, or the problems caused by large internet outages. And they tend to view the larger world through this sort of lens as being very frail. Many (perhaps even most) of these Tech CEOs that are discussing the potential apocalypticism of an AI powered world also have spent millions on their own survivalist security blankets. Even more is being spent on efforts to allow for an escape into space or to create protected utopias. They often place this perceived frailty of their own experience on to the rest of the human experience. I take most of this with a very large grain of salt. The real risks of AI in the present are not apocalyptic in nature - it is that AI can tell falsehoods so convincingly that it becomes easier and easier to manipulate people. And this is part of why AI is that much worse than when we were generating our lies all on our own.

[Dario_M]

9 hours ago, california boy said:

Did you just make up this statistic?

That was also my thought. 

Edited June 1, 2023 by Dario_M

[Kenngo1969]

14 hours ago, Benjamin McGuire said:

While I can see a chatbot being helpful in some ways, its also going to be really problematic in the current state of chatbot AI. I was reading a bit this afternoon about a federal judge in Texas who just banned arguments written entirely by generative AI. Why? An attorney used ChatGPT to supplement his argument, and he didn't check the cases that it cited. ChatGPT invented some of them entirely. ...

If I'm thinking of the same instance to which you are referring, the attorney "checked" the "cases," all right ... by asking ChatGPT the equivalent of the following: "Is Smith v. Jones [presumably, also, a citation was included] a real case?"  (Some people are getting awfully lazy when it comes to "earning" their $250 an hour ...    And to that poor, presumably-overworked attorney, Westlaw is your friend ... )

Edited June 1, 2023 by Kenngo1969

[Kenngo1969]

Quote

Me: Why was polygyny ended in the Church?  [I didn't feel the need to use the Church's full name, since, presumably, LDSBot knows from whence it springs]

 

 

Quote

 

LDSBot: Polygyny, a form of polygamy where a man has multiple wives, was practiced by some early members of the Church of Jesus Christ of Latter-day Saints. It was a part of the larger practice of plural marriage, which was believed to be a commandment from God at the time. However, the practice of polygyny and plural marriage ended in the Church for several reasons:

1. Legal pressure: The United States government enacted laws targeting polygamy and threatened to seize the Church's property and assets, as well as imprison those found practicing polygamy.
2. Statehood for Utah: The federal government made it clear that Utah wouldn't be granted statehood as long as polygamy continued to be practiced.
3. Safety and well-being of the Saints: The legal issues around polygamy caused significant persecution, making it difficult for the early Latter-day Saints to live safely and openly practice their faith.

In 1890, the Church's President, Wilford Woodruff, issued the Manifesto (Official Declaration 1), which stated that the Church would no longer practice plural marriage. This was a significant change and the result of revelation from God, in response to the challenges the Church was facing. Since then, the Church has continued to follow the guidance in the Manifesto and no longer practices polygyny or plural marriage. 😊

 

 

 

 

[Amulek]

7 hours ago, Benjamin McGuire said:

While I am aware of the fact that lots of people already make stuff up without the use of generative AI, part of the problem (as I see it) is that few people understand how generative AI works - and are more willing to believe what it tells them without the proverbial grain of salt.

This is certainly a concern. As a real life example, here's a story about an attorney who used ChatGPT to draft a motion. The AI returned exactly the case law the attorney wanted - because it made up the cases, the citations, and even the quotes.

Who know that AI could pose an existential threat to your law license.  

 

Edited June 1, 2023 by Amulek

[smac97]

6 minutes ago, Amulek said:

This is certainly a concern. As a real life example, here's a story about an attorney you used ChatGPT to draft a motion. The AI returned exactly the case law the attorney wanted - because it made up the cases, the citations, and even the quotes.

Who know that AI could pose an existential threat to your law license.  

A few months ago I asked ChatGPT to tell me the "Right of Redemption" period for tax foreclosure sales in Michigan.  Chat GPT quoted, chapter and verse, text that did not exist in the chapter and verse of the Michigan Code.  I pointed that out to ChatGPT, and it apologized and referred me to a different part of the Michigan Code, with the same result (phantom text).  After four times of this, ChatGPT apologized and said it did not have an answer.  I asked it where it obtained the text it had been quoted, and it said it did not know.

It was a pretty surreal experience.

Thanks,

-Smac

[MiserereNobis]

I had my creative writing students input one of their poems and ask chatGPT to analyze it. They really enjoyed seeing what AI thought their poems meant. I had them ask follow up questions about specific symbols, metaphors, etc. They then tweaked their poems based on that feedback. It was an enjoyable activity for all.

[mfbukowski]

23 hours ago, smac97 said:

Deseret News: Can we use AI to answer questions about faith?

I think attended is needed, but apocalyptic alarmism is not.

A few comparisons between this LDS Bot and ChatGPT:

1. "Was Martin Harris a credible witness for the Book of Mormon?"

ChatGPT Response:

LDSBot Response: 

2. "What evidence is there for the Book of Mormon being a translation of an authentically ancient text?"

ChatGPT Response:

LDSBot Response: 

3. "What is the best evidence against the Book of Mormon being a translation of an authentically ancient text?"

ChatGPT:

LDSBot:

Thoughts?

Thanks,

-Smac

Holy cow.

Stick the fork in.  Humanity is no longer relevant. We're done.

[mfbukowski]

52 minutes ago, smac97 said:

A few months ago I asked ChatGPT to tell me the "Right of Redemption" period for tax foreclosure sales in Michigan.  Chat GPT quoted, chapter and verse, text that did not exist in the chapter and verse of the Michigan Code.  I pointed that out to ChatGPT, and it apologized and referred me to a different part of the Michigan Code, with the same result (phantom text).  After four times of this, ChatGPT apologized and said it did not have an answer.  I asked it where it obtained the text it had been quoted, and it said it did not know.

It was a pretty surreal experience.

Thanks,

-Smac

On second thought....