The National Association of Evangelicals wrote down what must remain human

On October 1 the National Association of Evangelicals published a resolution on artificial intelligence. It is mostly a list of limits: what AI must never replace, and what no gain in productivity is allowed to cost. It is worth reading next to whatever AI policy your own organization has written.

The story

The National Association of Evangelicals wrote down what must remain human

An association representing 40 evangelical denominations has published a list of things AI must never be allowed to do. The National Association of Evangelicals has represented evangelical churches and organizations since 1942. Its resolution on artificial intelligence says at the top that it is not a final word on the subject, only a starting point, and it grants that AI can do real good in the world. Then it names the limits.

The line aimed most directly at business is about work. The resolution says that "gains in productivity should not be purchased by discarding the worker made in God's image."

I (Jason) am working on an essay based on one of the Hebrew harvest laws, described in Leviticus 19. Israelite farmers were told, in the words of the law, "do not reap to the very edges of your field." Whatever grew at the margins, and whatever fell during the harvest, was left for the poor and for foreigners who owned no field. The farmer gave up part of the yield on purpose, every year, as a standing cost of farming. A company that takes every available efficiency out of a process has reaped to the edges. The decision a leader can make now, before the savings arrive, is what share of them goes to the people whose work changed.

More this week

Americans are about as likely to take a hard problem to AI as to a religious leader

Asked where they would go with a difficult personal problem, 26 percent of Americans named an AI tool and 28 percent named a religious leader. NPR and Ipsos surveyed 1,011 American adults from August 7 to 9, with a margin of error of 3.6 points.

Among Americans who use AI at all, 69 percent say they would never ask it for religious or spiritual guidance, and 75 percent trust advice from people they know over advice from AI. Even so, 11 percent of AI users have already asked AI for spiritual or religious guidance, 13 percent have had it write a prayer, and 15 percent have used it to study scripture.

Why it matters for you: one reason people ask a model instead of a person is that the model will not think less of them for asking. If somebody on your team would sooner describe a mistake to a chatbot than to you, the thing to examine is how you react when people bring you bad news. Think back to the last three times someone brought you something you did not want to hear, and ask what the first words out of your mouth were. Then tell your team what you want them to bring you early, and what will happen when they do.

The practice

Do spiritual practices mix with your work-life?

At a conference on AI at Calvin University this month, the church track asked which spiritual practices "mitigate or counteract the negative effects of AI." Wisdom in the Age of AI ran from October 8 to 10 in Grand Rapids, hosted by the Calvin Center for Faith and Technology with the CCCU and INCHE, with tracks on business, education, health care, the arts, STEM and the church. Andy Crouch gave the opening keynote.

That question above sounds like one for a chaplain rather than a chief executive. Prayer, rest, silence and long-form reading were all built for people with limits. They are also the first habits to go when a capable tool is available every hour of the day. But a leader who keeps one or more of these models for their staff three things: that people are shaped by what they do over and over, that some friction is worth keeping, and that a slower pace is sometimes the right call.

So pick one or more, keep it/them for four weeks, and don't be bashful about it. Emphasize to your staff the importance of deep thought, patient learning, and taking time away from the grind.

Any paid AI plan (Claude, ChatGPT or Gemini) can tell you which commitments the practice will run into. Paste in the last two weeks of your calendar and ask:

Higher ground

The National Association of Evangelicals wrote a statement, and a statement is a decision made once. The letter to the Hebrews says that the people who can tell good from evil are the ones trained by constant use. Constant use means the same decision made again on an ordinary day, when nobody is watching and it would be easier not to. Write down what AI will not do in your organization. Then tell your team the one practice you are keeping while you lead them through it.

The next step

Not everybody should hand AI the same work, because people are not strong at the same things. Our free AI Edge assessment is 21 questions and takes about four minutes. It names your strongest of seven archetypes, then tells you what AI absorbs and what it amplifies for somebody wired that way.

Find your AI Edge →

Then hit reply and tell us which practice you picked. We read every one.

Climbing with you,
Jeff and Jason