Dealing with Emerging Technology
In my last blog, AI for Churches – A Pivotal Moment, I touched on how the pandemic felt like a bit of an unavoidable *shove* for many churches into the digital age, whether they were ready or not. Resource platforms like the Church of England’s Digital Labs newsletter and blogs became lifelines, helping churches launch livestreams, host Zoom services, and build online communities.
Now, what does this have to do with AI? My point is that the pandemic helped highlight two crucial areas:
Our churches have wildly varying levels of digital confidence and competence
When needed, we can respond and adapt to emerging technologies, and in some cases, pretty quickly!
On a blog all about AI, I’m not trying to say AI is another pandemic, even if some view it in such a light, but I am saying it’s a fast-moving wave that churches can’t ignore. I used ‘wave’ as a nod to an excellent book on the topic The Coming Wave, by Mustafa Suleyman. With these powerful tools now at everyone’s fingertips, with the likes of ChatGPT clocking almost half a billion users a week¹, we need to think carefully – and urgently – about how we, as a church, can use them.
Now, before your eyes glaze over at the word policy – yes, I get it. Some people love a good policy, and some people absolutely loathe them. Now as I’ve stated above, I think it’s essential that as a church, and wider society in general, we NEED policies on AI usage, but there is a balance to be struck. We don’t want to strangle ourselves out of using these tools to better our mission and ministry, but we do need clear, discerned guardrails in place with appropriate training and resources available for those who wish to use them. I’ll be discussing the ethics of AI usage in a future blog.
Why your Church needs an AI Policy
I think it’s really simple. It’s pretty likely that AI is already being used in your church in some way. The trendy ‘AI’ bumper sticker that’s now plastered everywhere as this new technology is really just advancements in the same tech we’ve been using for years. Live transcriptions on YouTube, blurring your background on Zoom — all of these little perks we take for granted come from machine learning algorithms.
Now, companies like OpenAI, Microsoft, Google, and Canva have tweaked and optimised their platforms to make complex AI tools accessible to everyone. And the kicker? They’re free and very easy to use.
Much like social media platforms, AI’s big selling point is convenience. You can jump on, ask your questions, tweak your documents — all without really thinking about the ethical implications or privacy risks. We should’ve learned this lesson after the Cambridge Analytica scandal in 2016²: if the product is free, YOU are the product.
Take ChatGPT. Sure, it’s a language model — but where do you think that language comes from? Or look at GrokAI, trained on millions of tweets, posts, and comments scraped from Twitter/X. Some people reckon this is why Elon Musk bought Twitter in the first place. The recent merging of X with xAI shows just how valuable that data really is.
Now, back to the question: why does your church need guardrails? Here’s a little simple breakdown:
1. AI is Already Shaping Your Voice (Whether You Know It or Not)
Every time someone uses ChatGPT to draft a sermon, or uses Canva to generate an event post, your church’s voice is being subtly shaped by those tools – tools that are trained on secular and worldly thinking and writing. For some, this might be a non-issue, but there’s a reasonable position that worries about the erosion of the value of ‘human’ input. People are using these tools to improve how they come across and are starting to lean on AI to “sound right,” but over time, you risk sounding like everyone else. Additionally, there’s a potential loss of identity in here.
Trembling FAR down the rabbit hole, one should also consider how this will effect communication in the long run – if everyone is receiving AI scripted messages everywhere they go, there will be AI models that reduce these messages to a language that you want to read. Imagine conducting a sermon to a congregation that would rather rely on the AI summary. As Christians we have a mission with a message to say. AI should support that, not replace it.
2. AI Doesn’t Know Your Context
AI pulls from everything. Good theology, bad theology, random blog posts, Reddit arguments – of which I’ve been in a few! In Search engine optimisation, Google ranks content based on three key factors; experience, authority, and trust, also known as E-A-T. I believe we must approach AI in a similar sort of manner when it comes to the sources of information. We need to make sure that if we’re going to use these tools, we do so carefully, and where possible ensure that we know the source material.
Some AI tools like NotebookLM allow you to create AI ‘chatbots’ based purely on material you have uploaded, which is a much safer method to controlling the source content. Ultimately, if you don’t have clear guidance, people might take AI-generated content at face value, not realising it’s built on a mishmash of sources with no filter for your church’s beliefs or values. It’s on us to use discernment here.
3. Privacy Isn’t a Suggestion
This one’s simple, as highlighted before. If someone pastes sensitive pastoral information into a public AI tool, like ChatGPT, that data is going somewhere. Even the paid ‘Plus’ version of ChatGPT uses your information to help train their models. This doesn’t mean someone will read that John Smith requested a prayer for his dog, but it’s still essential we remain sensitive to these issues.
Most AI platforms will tell you they “may use your data for training.” If you’re not setting boundaries, it’s too easy for someone to accidentally share something that should’ve stayed confidential. A policy helps people stop and think before hitting submit and forces us to at least have the conversation about what we deem as appropriate usage.
4. Clarity Beats Guesswork
Without a clear AI policy, we’re ultimately just a little lost with how to use these tools safely and effectively. What’s okay? What’s not? A simple, well-thought-out policy takes away the guesswork. It gives people confidence to use these tools well and with integrity. I know this is just the argument for why to have a policy for anything, but it’s a solid point. Understanding the sources, discussing the ethical/spiritual/environment implications, outlining risk assessments for varying levels of use. These are all extremely important considerations and again highlight the need for discernment and a clear policy in place.