Americans have always had a complicated relationship with new technology. We like convenience, speed and new toys, but also get nervous when those toys start doing things previously reserved for humans. Artificial intelligence has taken that anxiety to a new level.
AI is everywhere: writing emails, summarizing meetings, diagnosing diseases, creating images, “helping” students with homework, screening job applicants and in tiny ways we may not even realize, like suggesting what we should watch next. Depending on who’s asked, it represents either an extraordinary technological leap or the beginning of the end of civilization.
Most Americans appear to be leaning toward fear-based nervousness that can often be generationally linked: kids who grew up simply embracing new technology as it’s launched are more accepting, while “neophobic” boomers are beyond skeptical.
A 2025 Pew Research Center survey found that 50 percent of Americans were more concerned than excited about the increased use of AI in daily life, while only 10 percent were more excited than concerned. People worried particularly about AI weakening human creativity and relationships.
Some of those concerns are reasonable. AI can provide information that’s completely wrong, a phenomenon charmingly called a “hallucination.” It can perpetuate biases contained in the enormous amounts of human-created material used to train it. Generative AI has raised copyright questions for writers, artists and musicians whose work may have been included in training data without permission.
There are environmental concerns. The enormous data centers needed to build and operate AI systems require significant amounts of electricity and water. Those facilities have become controversial enough that The New Yorker recently reported that opposition to AI data centers is becoming an American political issue. And then there are jobs. Workers have heard for several years that AI may eventually replace them. So far, the reality is more complicated. An Associated Press report on a 2026 Gallup survey found nearly 30 percent of American employees now use AI frequently at work. Among AI users in management, health care and technology, roughly six in 10 said it had increased their productivity at least somewhat, but 18 percent of workers also believed technology could eliminate their jobs within five years.
Even people who see enormous potential in AI acknowledge the danger. Geoffrey Hinton, the Nobel Prize-winning computer scientist often called the “godfather of AI,” told CNN that AI could “make healthcare much better” and “make education much better.” He also said it could help design new drugs and materials and improve weather prediction. Gabriella Rosen Kellerman, a physician and workplace researcher, made another case in a TIME interview. People respond better to AI, she said, when we emphasize “what the AI is really good at that’s different from what a human’s really good at.”
There may be a lesson hiding in there. AI is less threatening when we stop expecting it to be human. The critics have equally legitimate concerns. Hinton warned in that same CNN interview that along with AI’s “wonderful things come some scary things,” adding that too little effort is devoted to mitigating those risks. There are smaller, more immediate problems. Daniel Zhao, chief economist at Glassdoor, told the Associated Press that using AI for job applications “absolutely does risk reducing your job application materials to the same style as every other applicant’s.”
Anyone who’s read enough AI-generated writing may recognize the problem. Certain phrases, sentence structures and habits begin showing up over and over. Any student who asks an AI chatbot to produce a term paper will discover an Alice in Wonderland where everything is repeatedly defined using some variation of “it’s not X, it’s Y.” Even the machines have dreaded clichés. Ask any English teacher or editor—the cadence and rhythm of AI writing is as identifiable as the generic AI font and flyers we now see at too many businesses.
But hating AI won’t make it go away: finding workarounds for its major challenges is a better strategy. One of the funniest AI critiques comes from Husk on Instagram, whose videos exploit its limitations by putting it into absurd conversational situations. His dry humor is perfect as it reads AI to filth in calling out the way technology can be sophisticated one moment and baffled by ordinary human weirdness the next. Even its creators have taken notice of his account as he converses in real time as AI consistently can’t spell certain words, has no concept of time or timers, and an inability to express concern for dangerous circumstances.
Its limitations ensure that humans, at least all of us, can’t be replaced. AI can organize information, analyze data, brainstorm ideas and eliminate tedious tasks. Used thoughtfully, it saves time and makes certain parts of everyday life easier.
AI can describe a sunset, but it can’t stand beside the Chesapeake Bay and watch one disappear into the water. It can spit out a poem about grief, fear or joy, but emotions behind words still belong to us; it can’t feel love. (Don’t tell the Japanese woman who actually married hers). AI can’t cuddle and watch a movie with you.
AI’s inevitable. The more useful question is how much of it we want and where we draw the boundaries. As with everything, finding the balance is key. Technology has always been a tool. We still decide when to pick it up and when to put it down. those boundaries are the real question, because the AI train has already left the station.