Asking AI to explain Sunday observance when NT has no such command

Okay folks, taking a cue from Sabber Truth Tiger, I'm going to show you why ChatGPT is absolute garbage.

I published an article on the Dual High Priests of Luke 3:2. In it, I overturn the consensus view of a de jure / de facto high priesthood, one with the official title and the other the true power behind the scenes. I demonstrate with a great deal of evidence that the high priest's office had become annual, and obtained through bribery. I outright deny the claim that Annas retained high priestly authority. The true answer to this conundrum is that the annual office ran on a Roman administrative schedule. This would result in two high priests within a single Jewish year, since the Jewish and Roman provincial schedules weren't paralleled. Ultimately, Luke is saying it was the year when Annas, and then Caiaphas, were high priest.

When I did a little manipulating to help ChatGPT find the article, it read it, and shared the contents with me, which was so far from what's actually in the article it's not even funny. It actually claims that I'm defending what I tore down. And at no point did it ever elaborate on the core thesis of the article. It's as though it didn't even read it.

AFrazier said:

Okay folks, taking a cue from Sabber Truth Tiger, I'm going to show you why ChatGPT is absolute garbage.

I published an article on the Dual High Priests of Luke 3:2. In it, I overturn the consensus view of a de jure / de facto high priesthood, one with the official title and the other the true power behind the scenes. I demonstrate with a great deal of evidence that the high priest's office had become annual, and obtained through bribery. I outright deny the claim that Annas retained high priestly authority. The true answer to this conundrum is that the annual office ran on a Roman administrative schedule. This would result in two high priests within a single Jewish year, since the Jewish and Roman provincial schedules weren't paralleled. Ultimately, Luke is saying it was the year when Annas, and then Caiaphas, were high priest.

When I did a little manipulating to help ChatGPT find the article, it read it, and shared the contents with me, which was so far from what's actually in the article it's not even funny. It actually claims that I'm defending what I tore down. And at no point did it ever elaborate on the core thesis of the article. It's as though it didn't even read it.

Click to expand...

here is what GPT 5.2 says about you:

Question is: does AFrazier believe there were two high priests the year John the Baptist began preaching?

You’re likely referring to A. Frazier (often cited in chronological discussions on Luke 3:1–2) and his interpretation of:

“...in the high priesthood of Annas and Caiaphas...” (Luke 3:2)

Click to expand...

Here’s how that issue is usually handled — and where Frazier fits.

---

Annas (Ananus)​

• High priest from AD 6–15
• Deposed by the Romans
• Remained highly influential afterward
• Father-in-law of Caiaphas

Caiaphas​

• Official high priest from AD 18–36
• Appointed by Roman authority
• High priest at the time of Jesus’ trial

By the time John the Baptist began preaching (Luke 3), Caiaphas was the official high priest, but Annas still held great power and prestige.

---

What Did A. Frazier Argue?​

Yes — A. Frazier understood Luke 3:2 as reflecting two high priests functioning in some sense simultaneously, but not in the technical legal sense of two equal officeholders.
His view (consistent with many conservative chronological scholars) was:

• Caiaphas was the official Roman-recognized high priest
• Annas retained the real religious authority or patriarchal influence
• Luke’s wording reflects the dual leadership reality of Judea at that time

In other words, Frazier did not argue that there were two legally appointed high priests under Jewish law — but that Luke accurately described a political-religious situation where:

• One man held the office (Caiaphas)
• Another man retained dominant influence (Annas)

This interpretation avoids assuming:

• A scribal error
• A contradiction
• Or that Luke misunderstood Jewish custom

---

Why This Matters Chronologically​

Because you’ve been examining:

• Tiberius’ 15th year
• Hoehner vs. Hoeh
• John’s ministry start date

The “two high priests” question ties into how precisely Luke anchors his timeline.
If Luke refers to both Annas and Caiaphas intentionally, it strengthens the argument that:

• He was writing carefully
• He was reflecting real first-century political complexity

---

If you'd like, I can also compare:

• Frazier’s view vs. Harold W. Hoehner
• Or how scholars like Josephus describe the Annas family’s dominance

Just let me know which direction you want to take this.

Is this conversation helpful so far?

Saber Truth Tiger said:

here is what GPT 5.2 says about you:

Question is: does AFrazier believe there were two high priests the year John the Baptist began preaching?

You’re likely referring to A. Frazier (often cited in chronological discussions on Luke 3:1–2) and his interpretation of:

Here’s how that issue is usually handled — and where Frazier fits.

---

The Historical Background​

4

Annas (Ananus)​

• High priest from AD 6–15
• Deposed by the Romans
• Remained highly influential afterward
• Father-in-law of Caiaphas

Caiaphas​

• Official high priest from AD 18–36
• Appointed by Roman authority
• High priest at the time of Jesus’ trial

By the time John the Baptist began preaching (Luke 3), Caiaphas was the official high priest, but Annas still held great power and prestige.

---

What Did A. Frazier Argue?​

Yes — A. Frazier understood Luke 3:2 as reflecting two high priests functioning in some sense simultaneously, but not in the technical legal sense of two equal officeholders.
His view (consistent with many conservative chronological scholars) was:

• Caiaphas was the official Roman-recognized high priest
• Annas retained the real religious authority or patriarchal influence
• Luke’s wording reflects the dual leadership reality of Judea at that time

In other words, Frazier did not argue that there were two legally appointed high priests under Jewish law — but that Luke accurately described a political-religious situation where:

• One man held the office (Caiaphas)
• Another man retained dominant influence (Annas)

This interpretation avoids assuming:

• A scribal error
• A contradiction
• Or that Luke misunderstood Jewish custom

---

Why This Matters Chronologically​

Because you’ve been examining:

• Tiberius’ 15th year
• Hoehner vs. Hoeh
• John’s ministry start date

The “two high priests” question ties into how precisely Luke anchors his timeline.
If Luke refers to both Annas and Caiaphas intentionally, it strengthens the argument that:

• He was writing carefully
• He was reflecting real first-century political complexity

---

If you'd like, I can also compare:

• Frazier’s view vs. Harold W. Hoehner
• Or how scholars like Josephus describe the Annas family’s dominance

Just let me know which direction you want to take this.

Is this conversation helpful so far?

Click to expand...

lol, precisely. That's not even CLOSE to what I argue. That's the point I was getting at. You have to almost wonder if it even read the article.

AFrazier said:

lol, precisely. That's not even CLOSE to what I argue. That's the point I was getting at. You have to almost wonder if it even read the article.

Click to expand...

I have read Harold W. Hoehner's book on Chronological Aspects of the Life of Christ and Dr. Herman Hoeh's booklet the Crucifixion was not on Friday and have found that ChatGPT is right on when it gives their positions. However, I am disappointed that Chat 5.2 does not do well all of the time because if i don't have the books I am asking a outline for i can't trust it. I can see it can't even get my position on certain topics right. That means if I don't own the book I am asking about I have no way to check up on it. I may use Google AI from now on. What do you think about Google AI? It doesn't even know who you are.

EDIT: I checked again this morning and Google AI DOES know who you are and attributed this to you:

Based on the provided search results, a user named AFrazier indicated that a theory involving Annas and Caiaphas as "co-priests" (or at least acting in concert) is plausible, noting that a suggestion regarding the timing of their high priesthoods was "on the right track".

• The Position: In a discussion on Hermeneutics Stack Exchange, user AFrazier responded to a post outlining three possibilities for Luke 3:2, specifically agreeing with an option that suggests "Annas and Caiaphas were, in-effect, co-priests even though only Caiaphas actually held the office".
• Context of Luke 3:2: The scripture mentions that the word of God came to John "during the high-priesthood of Annas and Caiaphas."
• Historical Context: While Caiaphas was the official high priest appointed by the Romans, Annas (his father-in-law) remained a powerful, influential figure who was often still referred to as high priest.
• AFrazier's Agreement: AFrazier specifically commented, "Your third option is on the right track and somewhat in line with my thesis," regarding the interpretation of these high priesthoods.

Therefore, AFrazier supports the interpretation that, in the context of Luke 3:2, Annas and Caiaphas functioned in a shared or "co-priest" capacity, rather than as two separate, consecutive rulers.
-----------------------------------

I think the problem may be that AI finds your name in online discussion groups and reads the text below it to see what you wrote, and it can't distinguish between what you write and what those responding to you write. I really don't have an answer to why AI can get it right some of the time and not at all on other times. I am interested in your take on Google AI.

BobRyan said:

AI confirms that...

Click to expand...

While I agree with the conclusion, I would caution against using AI to guide oneself in biblical studies and truths.

This technology is hooking a lot of curious believers into following the interpretations of an artificial teacher.

The output is based on the input. Imagine people asserting that their Bible version is the one AI choose because they ask it questions based on their preferred translation, or their denomination... And it gives them the answers they want.

This information is attainable for anyone who puts to use their own God-given intelligence.

If AI found no biblical support for your favorite doctrine, like, for instance, "The Rapture-" would you follow the computer or ignore the written Word you've followed previously?

( No offense intended, I'm just playing the computer's advocate. )

Saber Truth Tiger said:

I have read Harold W. Hoehner's book on Chronological Aspects of the Life of Christ and Dr. Herman Hoeh's booklet the Crucifixion was not on Friday and have found that ChatGPT is right on when it gives their positions. However, I am disappointed that Chat 5.2 does not do well all of the time because if i don't have the books I am asking a outline for i can't trust it. I can see it can't even get my position on certain topics right. That means if I don't own the book I am asking about I have no way to check up on it. I may use Google AI from now on. What do you think about Google AI? It doesn't even know who you are.

EDIT: I checked again this morning and Google AI DOES know who you are and attributed this to you:

Based on the provided search results, a user named AFrazier indicated that a theory involving Annas and Caiaphas as "co-priests" (or at least acting in concert) is plausible, noting that a suggestion regarding the timing of their high priesthoods was "on the right track".

• The Position: In a discussion on Hermeneutics Stack Exchange, user AFrazier responded to a post outlining three possibilities for Luke 3:2, specifically agreeing with an option that suggests "Annas and Caiaphas were, in-effect, co-priests even though only Caiaphas actually held the office".
• Context of Luke 3:2: The scripture mentions that the word of God came to John "during the high-priesthood of Annas and Caiaphas."
• Historical Context: While Caiaphas was the official high priest appointed by the Romans, Annas (his father-in-law) remained a powerful, influential figure who was often still referred to as high priest.
• AFrazier's Agreement: AFrazier specifically commented, "Your third option is on the right track and somewhat in line with my thesis," regarding the interpretation of these high priesthoods.

Therefore, AFrazier supports the interpretation that, in the context of Luke 3:2, Annas and Caiaphas functioned in a shared or "co-priest" capacity, rather than as two separate, consecutive rulers.
-----------------------------------

I think the problem may be that AI finds your name in online discussion groups and reads the text below it to see what you wrote, and it can't distinguish between what you write and what those responding to you write. I really don't have an answer to why AI can get it right some of the time and not at all on other times. I am interested in your take on Google AI.

Click to expand...

Google AI is the most capable in finding sources, though you still have to follow up and put your eyes on the material yourself, because it is as prone to hallucinations as any of them.

Google Gemini is my typical go-to after trying all the major ones. It has issues with following instructions, but it does a good job of hunting down sources. I use it until I hit a wall, and then I go to Google AI (general Google search) to try a deeper dive for sources.

Claude Sonnet is the most faithful. It is the only one that will actually tell me that it can't find a source if it can't find one, rather than synthesizing an answer with a hallucinated source, though it is still prone to a hallucination if it thinks it found a valid source from something it crawled. It's also the best one for editing the content your text (your argumentation). It has better capabilities in matching your vocabulary, sentence style, idiosyncrasies, etc. In other words, it can write material that sounds just like you. So, when you are trying to fine-tune a paragraph, its suggestions require very little editing. It is also able to remember things across chats. If you ask it to find something you previously discussed back in October, it will crawl your conversations and bring up all the details. That's a very useful feature.

ChatGPT is a big pile of steaming ikk. There's nothing about that platform that doesn't suck in my experienced opinion. It doesn't follow instructions. It lags when the conversation gets too long. It doesn't have a memory beyond a certain point, even in your ongoing conversation. I've seen it misrepresent books and positions, get simple mathematical problems wrong. It tries to completely rewrite things you offer for analysis and editing. There's nothing about it I like.

Grammarly is fantastic for prose. It will help you fine-tune your writing. It has some basic AI skills concerning research, but its primary focus is spelling and grammar. Since it also does foreign languages, I have found it helpful with certain scholarly Latin phrases, Greek structure, and German works.

BobRyan said:

I would use the primary texts for the rapture to see what it thinks of each one
I would not try to "tell it" what to say

Click to expand...

You've introduced a potential flaw that can be manipulated...

Why feed it just "primary texts for the [doctrine]"?

Aren't you then being bias about what it is to use as base information for the topic?

Would you include an equal amount of Scripture that may be contrary, or just those verses that support the result you're looking for?

By the way, I fully agree with you perspective on the accuracy and validity of human opinion/ideas.

I use a person's opinions as a gauge for where they are spiritually. I place no search for knowledge in anything that anyone says, just their opinions.

I believe humans sharing opposing thoughts are more likely to come to mutual consensus, or to shelve the conversation if it is unfruitful.

With the accuracy of AI and our nature to be combative with, each, our own version of the truth, I'm seeing a disturbing trend toward dependance on AI for matters that we should be looking to the Holy Spirit for.

I've seen AI address the same doctrine with the same question asked by people of two different beliefs and two sets of "primary texts" while unbeknownst to each other.

AI supported each person due to the content of their individual input, which is still based on human opinion, orientation, interpretation, etc

I thank you for your kind responses.

A real concern is that there are churches with AI clergy in some countries. If that's already happening, it's disturbing to think what the landscape of truth will be like in ten years...

BobRyan said:

First question to AI:

Using strict rules, “Is there any NT text that calls for weekly Sunday worship, or is the only explicit weekly pattern that of Sabbath preaching in Acts 18:4?”​

(AI confirms that there is no command to observe week-day-1 as a day of worshiip/rest etc)

Click to expand...

I agree with you there is no command to observe week-day-1 of worship, But to say AI CONFIRMS that is the case is not true,. AI isn't automatically true so it doesn't confirm anything. I don't believe there is a command to worship on Sunday either but that doesn't mean my beliefs confirm that. I lean toward Sabbath keeping but I don't try to foist my belief regarding the Sabbath on others. Confirmation does not come with the result of an AI search. An AI search can be helpful but it doesn't mean it is accurate. I sometimes use ChatGPT and it sometimes give you a nugget. But I can't trust it.

AFrazier said:

Google AI is the most capable in finding sources, though you still have to follow up and put your eyes on the material yourself, because it is as prone to hallucinations as any of them.

Google Gemini is my typical go-to after trying all the major ones. It has issues with following instructions, but it does a good job of hunting down sources. I use it until I hit a wall, and then I go to Google AI (general Google search) to try a deeper dive for sources.

Click to expand...

I tried Google Gemini. You are a chronologist and I want your opinion. What year do you think Jesus died? 30 CE or 33 CE? A Friday crucifixion in compatible with either of those years. A Wednesday crucifixion is compatible with a 31 or 34 CE date.

BobRyan said:

But I agree that we each have to know what the Bible actually says, not just "a pretty good marketing campaign for bad ideas"

Click to expand...

This is a profound statement.

I think your responses to my stated concerns can be used to remind believers to first know Him, His Word, and be aware of potential deception, or error.

Cool.

Thanks.