From the Editor: Neck and Neck

My favorite music station plays an unpredictable mix of current experimental pop, with older tunes across genres and decades. A woman’s voice periodically breaks in, inviting listeners to stay tuned: “Stay with us—and take a break from the algorithm,” she declares. I’m glad to oblige. I feel momentarily freed from the sleepless synthetic AI maestros that generate playlists they predict I’ll like according to the latest data sweep. 

If AI is a mystery to us, maybe it’s because we are still a mystery to ourselves.

But hold on. Wait a minute. Is the woman real? “Take a break from the algorithm” … Could that phrase and that voice be the crafty, flattering machination of an algorithm? Should I contact the station and inquire, or just roll with it?

Such questions are causing vertigo these days—what’s real, what’s fake, and does it even matter as long as it works? As we enter a new stage of AI-enhanced medical breakthroughs, economic efficiencies, and time-saving practices as well as AI-generated deepfakes, reputational ruin, crippling propaganda, and catastrophic fraud, yes, the real still matters. 

It’s thorny to assess this moment at all. We seem poised at a historic crossroad where we either allow AI to enfold itself quietly and exponentially into our lives full-on, or we strenuously object to further incursions. Or both! It’s neck and neck. In 2019, scientist Melanie Mitchell could write: “… the field of AI is in turmoil. Either a huge amount of progress has been made, or almost none at all. Either we are within spitting distance of ‘true’ AI, or it is centuries away. AI will solve our problems, put us all out of a job, destroy the human race or cheapen humanity. It’s either a noble quest or ‘summoning the demon.’”[1]

What’s different five years later is that we find far more extensive daily use of AI as well as a greater intensity of hopes and fears around its future. How to guide or control or manage it? We still haven’t found a way around Collingridge’s dilemma, which writer Mark Chinen in this issue deftly describes: “When a technology is in its early stages, it is too soon to regulate because it is impossible to predict what effects it will have, but by the time those effects are known, the technology is too widespread to harness.” 

Yet we better try. Around 2010, there was a vast civic unwillingness to understand or restrain the power of that new thing called social media—in splashy tandem with iPhones, high-speed internet, unlimited data plans, Instagram for tweens. Its effects now look like a desolation for many young people’s health and sense of worth. From the start, the market was left to its own devices. Chaos was embraced as progress. Revulsion against social media excesses arrived a decade later, if not a decade too late.

“The phone-based life produces spiritual degradation, not just in adolescents, but in all of us,” says social psychologist Jonathan Haidt, author of The Anxious Generation.

By contrast, the scale of AI will run larger and faster, with bigger economic impacts on whole populations. The national conversation is a growing mashup of excitement, wariness, ambition, naivete, skepticism, and alarm. For now, I keep seeing an odd word associated with descriptions of AI—it’s like “magic.” This becomes a default word for baffled delight, suggesting a state of fond, inarticulate wonderment. Yet the cold equations at work in Large Language Models are anything but magical. Something as real as AI shouldn’t be conveyed with language inspired by sleight-of-hand and trickery. 

Into the breach arrive these Reflections writers, whom I thank for daring to share their thoughts/anxieties/hopes about this fluid/brilliant/nerve-wracking period, when no one can be sure if the now-unfolding epic encounter between humans and AI is a sci-fi-style rom com, a world-bending collaboration, or the next world war.

Included in this issue, as always, is a poetry section. I love to discover and share new (or old) poems that stretch the meaning of the latest Reflections theme by their own eccentricity. The poets here can be observed observing. They can be heard listening—to the incomprehensible life of nature, to their own exhilaration, heartbreak, or mixed feelings. After a while, the online inscape feels claustrophobic, very indoors, a contrivance to make you forget there’s something bigger out there than AI will ever be: the earth, the cosmos, and our consciousness of those things. Poets notice shifts in personal spirit and national mood. They hold interiority up to the light. They ride an elusive emotion to the end of the line. They remind me that we are all globes of consciousness walking around, and consciousness is some sort of miracle, and every person’s consciousness is unique. To me, short poems especially drive this point home: they are photons of vehement experience, neuro-charges, soul-renderings. No lasting poem could be predicted or generated. It slides under the ideological radar. It chronicles contradiction and astonishment. Poems let their freak flag fly.

If AI is a mystery to us, maybe it’s because we are still a mystery to ourselves. I hope this Reflections honors what’s grandly and impossibly human about humans while demystifying the technologies we’re inviting into our futures and our destinies.

Reflections editor Ray Waddle is the author of “This Grand Errand”: A Bicentennial History of Yale Divinity School (YDS, 2022).