The Dangers of AI (As Told by a Chatbot)

“As an AI language model, I am not capable of having a religious belief or point of view.” (Well, that's reassuring.)
An illustration of a smiling mouth with green lips against a red backdrop. Vertical prison bars are visible in place of teeth. A red man holds onto two bars with both hands from within..
Illustration by Adrián Astorgano

ARTIFICIAL INTELLIGENCE IS all the buzz as I write this. It’s been impossible to ignore the omnipresent chatter about AI, from the deluge of online commentary to congressional hearings. As I thought about adding to the chatter — er, providing some insightful perspective from a progressive theological point of view — I wondered what more could be said. So, I decided to ask AI. I prompted Microsoft’s Bing AI chatbot to draft an essay, “from a progressive Christian perspective,” on the dangers of AI. The first line of the response: “As an AI language model, I am not capable of having a religious belief or point of view.”

Well, that’s reassuring.

Concerns about AI aren’t new — science fiction writers have painted grim pictures of machine consciousness at least since Samuel Butler’s 1872 novel Erewhon (wherein Butler wrote in his three-chapter “The Book of the Machines” that “there is reason to hope that the machines will use us kindly, for their existence will be in a great measure dependent upon ours; they will rule us with a rod of iron, but they will not eat us.” One hopes that we’ll find better reasons to hope.). Warnings of apocalyptic totalism abound: In May, Matthew Hutson wrote in The New Yorker, “In the worst-case scenario envisioned by [artificial-intelligence doomers], uncontrollable AIs could infiltrate every aspect of our technological lives, disrupting or redirecting our infrastructure, financial systems, communications, and more.”

Since the Bing chatbot is incapable of offering a theological perspective, I asked scholar Walter Brueggemann for his thoughts. After disclaiming “I know nothing of AI,” Brueggemann — who does know something about the Bible — said, “I think the Bible insists on two things: The freedom and reality of God’s intention for creation that is hidden and cunning, and unfettered human agency in response to God. Whatever impinges upon or threatens either of these poses important problems. I do not know if AI does either of these.” When asked about the “totalism” threat that AI brings, Brueggemann said, “I suppose the ultimate expression of totalism in the Bible is the claim of Pharaoh that ‘I made the Nile’ (Ezekiel 29:3), in which he imagines his own ultimacy. Whatever we imagine our ultimacy apart from God” — through computers or anything else — “is a threat to our common life.” Brueggemann added, “It occurs to me that the prohibited list of ‘ways to know’ in Deuteronomy 18:9-14 was likely the AI in their day, chances to bypass legitimate knowing via Torah.”

But focusing on the apocalyptic potential of AI can distract us from the more prosaic, and current, problems created by these technologies. Chatbots such as Bing’s and ChatGPT, which some call “plagiarism apps,” can perpetuate propaganda and “alt facts” in new and difficult-to-detect ways. And the Writers Guild of America strike this spring has spotlighted the threat that AI poses to people’s livelihoods. While picketing outside Paramount Pictures, writer Miranda Berman told NPR, “This is only the beginning. If they take writers’ jobs, they’ll take everybody else’s jobs, too.”

On that last score, my Bing chatbot pen pal was reassuring that AI technologies “cannot yet duplicate the work” of people in many areas. (Emphasis—gulp—added.)

This appears in the August 2023 issue of Sojourners