AI rhetoric swings between utopia and apocalypse. The likeliest outcome, though, may be far less spectacular: AI will give us exactly what we want, turning us into pampered infants living emptier lives, for whom effort itself becomes a kind of torture — a future less like Star Trek or The Terminator and more like WALL-E.
Laziness is adaptive: when rewards are equal, evolution favors organisms that conserve energy. Biologists call this the theory of least effort. We’re wired to minimize effort because organisms that didn’t waste calories on low-payoff activities had an edge over those who did. At the same time, animals that sat on their ass all day were outcompeted by animals that invested effort when it mattered — finding food, avoiding danger, securing mates, and raising offspring. This tension between saving energy and the struggle to reproduce came to a delicate balance: conserve energy whenever possible, but spend it when it counts.
The link between effort and survival created a simple rule of thumb: if something is hard, it’s probably worthwhile. To make that heuristic stick, evolution had to incentivize work, so it shaped a two-track reward system — immediate pleasure for conserving energy and chilling in the hot tub, and a deeper sense of satisfaction and meaning for doing hard things like earning a degree or training for a marathon. Effort is the wedge that separates pleasure from meaning — evolution’s solution to the problem of getting energy-conserving animals to do hard things when necessary.
At some point in our history, however, many humans — and some of our pets — crossed a threshold where effort became less a condition for staying alive and more a matter of choice. And once it becomes optional, the temptation isn’t just to avoid the tedious stuff — it’s to avoid doing anything at all.
The struggle that AI teaches us to avoid is often the very thing that makes life worthwhile.
I spent the morning copying the responses to a project I created with ChatGPT into another prompt to get “paste-ready” edits. My “job” is Control-A, Control-C, click on another tab, Control-V. Then I do something else — check email, watch a video on how to stop double faulting in tennis — while I wait for the LLM to spit out revisions and new files. Then I paste those back into the project tab. The only trick is keeping the order straight so you don’t have to read any of the junk (i.e. the comments generated by someone else’s LLM) that I’m feeding back into mine. They pay me $65 an hour to do this, and I’m guessing it’s only a matter of time before the company decides the obvious: cut out the middleman and let the machines talk to each other directly.
The job might actually be interesting if I bothered to read any of the responses or did any of the thinking myself. I’m supposedly training an LLM to do statistical analyses better, and I might actually learn something if I treated it like work instead of churning out deliverables. But I can log more hours if I just optimize it by never reading any of the comments my “teammates” paste in from their own LLMs.
None of this is entirely new, of course, and there were lots of mindless jobs before AI. The washing machine shrank laundry from a day-long ordeal into a quick chore, freeing millions of women for paid work — or anything besides scrubbing clothes by hand. Calculators and search engines did the same for mental work, supposedly saving us time to do “more important” tasks.
But AI is different. It doesn’t just eliminate drudgery; it increasingly competes with us in the higher-order cognitive tasks it frees us up to do. The danger is that it threatens the ancient bargain between pleasure and meaning by making effort avoidable not only for chores but for the struggles that produce competence and meaning.
Generative AI isn’t just another labor-saving device making us more efficient — it’s becoming the thinker. Three years after ChatGPT, generative AI can’t just write a student’s paper, it can complete entire online courses. The problem isn’t that it’s making tasks easier; it’s making them optional — and slowly convincing us that doing them at all is pointless.
The distinction between whether a technology complements our skills or substitutes for them is critical. Tools that complement our skills still leave the user doing the work, while tools that substitute for them destroy competence. As AI gets better and better at doing intellectual work — drafting emails, writing essays, summarizing vast troves of information — the tendency to label tasks “mindless” to justify skipping them becomes greater. Is reading a book, instead of skimming a summary of bullet points, a waste of your time? Is forcing yourself to wrestle an original sentence into existence mindless?
The more I use generative AI, the more I feel that the important things in life are being stolen from me. We are approaching a world where college essays are written and graded by LLMs with little human involvement. Students ask ChatGPT to write essays that their instructors ask ChatGPT to grade. It is deeply ironic that more and more kids (or their parents) are paying $60,000 a year to cheat their way to a degree, doing everything possible to minimize effort and learn as little as possible.
Immediate gratification and frictionless ease now characterize modern life. Our economy is built around one-click purchases, same-day delivery, and frictionless checkout lines. Consumers have become so twitchy, impulsive, and allergic to lag that adding 100 ms (0.001 seconds) for an order on Amazon was seen to reduce sales by approximately one percent (or $6.4 Billion) and the average American checks their phone once every five minutes.
What’s the point of it all? I’ve even heard people bragging about using AI to automate the online part of online dating. Isn’t the whole point of dating the enjoyment of meeting and learning about someone else? I don’t want to streamline my relationships for the same reason I don’t want a laugh track to do my laughing for me. Why are we trying to bypass fundamental elements of the human experience? When machines eliminate friction they strip away the meaning that comes from the struggle. ChatGPT will never teach me the way the real world does — through wrong turns, wasted time, embarrassment, and hard-earned mistakes.
AI didn’t start this, but it looks like it might complete it: the comfort trap, a world where everything is easy but nothing feels earned. I don’t blame people using clever tools to avoid genuinely menial work. The problem isn’t that we like ease; it’s that AI changes the incentives so the meaningful stuff becomes optional too.
Effort is what gives our lives meaning; it’s what separates living life from consuming it. Outsourcing effort makes tasks easier, but it turns experience into a checklist — where “seeing” all the paintings in the Winter Palace matters more than the slow pleasure of actually being there. Meaning is downstream of effort. If you want the meaning, you have to pay the effort tax.
The answer isn’t to reject every shortcut. It’s to recognize that there’s more to living a good life than putting as little effort as possible into every task and get back to a place where we aren’t only doing things in order to finish them. Read the book, not the summary. Sit with the blank page for a few minutes. The struggle that AI teaches us to avoid is often the very thing that makes life worthwhile.
The fantasy is that automation will buy us time for the good stuff — tennis, meditation, learning to play an instrument. But these things take time and effort, and more often than not, automating tasks just buys us time for more Candy Crush. The biggest risk of AI may be happening right under our noses: a life optimized for comfort, where anything hard gets outsourced, nothing feels earned, and life feels empty.
READ MORE:
How COVID Created the 15 Second Generation
Suing Social Media Won’t Save the Children — But It Could Silence Everyone
Robert Lynch is a data scientist and a bio-cultural anthropologist, specializing in how biology, the environment, and culture come together to shape health and behavior at Pennsylvania State University. His substack is The Laughing Ape Substack and his website is here.