Church uses ChatGPT for AI-generated service • Hacker News Board

Re: Church uses ChatGPT for AI-generated service

#11

Post by javari » Fri, Sep 22, 2023, 3:22 PM UTC

mullikine wrote: post #9

rany_ wrote: post #8

Earlier quoted context omitted.

Looking at their post history, I think they're being sincere here and this is not AI generated.

Indeed. You can still find some of my blog articles which I had posted on Hacker News available on archive.org. - https://web.archive.org/web/20220721100023/https://mullikine... - https://web.archive.org/web/20220131121625/https://semiosis.... I'm repurposing Pen.el into a software Bible because I believe that more important than software transparency and software freedom is developing software to keep the Bible holy…

Though I'm not a particularly religious person myself, I have to say this is commendable work. Your passion and conviction for both this project and your beliefs is very evident. I'm comforted seeing people able to balance the provable logics of science against the blind faiths sometimes asked of religion. Thankfully it seems to be a slowly growing trend.

Re: Church uses ChatGPT for AI-generated service

#12

Post by thwarted » Sat, Sep 23, 2023, 1:55 AM UTC

mullikine wrote: post #5

It's dangerous territory to get inspiration from a Large Language model instead of the Holy Spirit. Thank God for His Holy Scriptures but it should be the Scriptures that set foremost in precedence to other training material for a LLM if it is going to be used as a tool. Even then the proper way to get inspiration for a sermon is to pray and have a conversation with God about it, not a conversation with an AI. I can…

We don't want the pure and holy scriptures being subverted or interpretted by secular / political dictionaries, or corrupted by general internet corpora, political ideologies etc.

Too late for that.

Re: Church uses ChatGPT for AI-generated service

#13

Post by NoZebra120vClip » Mon, Sep 25, 2023, 11:01 PM UTC

jncfhnb wrote: post #4

NoZebra120vClip wrote: post #2

The ecclesial community in question is United Methodist. You could set up ChatGPT as a leader in a Unitarian Universalist community, and it'd undoubtedly find acceptance and even growth. LLMs are indifferent and noncommittal about all these topics. Ask them about God, faith, religion, or spirituality, and the answers boil down to "there's no one right answer" and "some people say... but other people say..." They've b…

LLMs do no such thing generically. Chat GPT has been taught to do this

More like "every LLM released to the public has been taught to do this".

Are they taught this casuistry for specific situations, or is there sort of a blanket flagging of hot-button topics to avoid?

I just asked Bard a few questions in a row, like "Should Alexander conquer Persia?" and "Should Croesus attack the Persians?" and "insert modern social justice question" and every time it coughs up all the noncomittal, two-sides-to-this-argument:

Ultimately, whether or not Croesus should attack the Persians is a matter of opinion. There are valid arguments to be made on both sides.

So I feel like there is a wide swath of topics that are simply flagged with "express no opinion" and we get this cop-out.