Technology
Outsourcing Thought What Handing Small Decisions to AI Does to Our Minds
From drafting texts to settling dinner debates, we now hand small decisions to machines. What do we gain and what quietly erodes?
Listen to the opening
0:003:19
In this episode
- The Last Time You Sat With Not Knowing
- An Old Bargain at a New Scale
- What a Brain Does When It Isn't Asked
- Trusting the Machine the Right Amount
- The Machine That Listens
- Deciding What to Hand Over
- What We Keep
The opening
When was the last time you were unsure about something small and simply stayed unsure? A friend asks whether a tomato counts as a fruit. Someone wonders how long to rest a roast, or whether "regards" or "best" is the right way to sign off to a colleague you barely know. A few years ago, those moments had a natural length. You guessed, you argued, you half remembered something, or you let it go. Today, for a growing share of people, the moment lasts about as long as it takes to open an app.
This is the story of that shift, and of the researchers trying to measure what it costs and what it gives. The early chatbot boom of 2024 felt like a novelty, a parlor trick that could write a limerick about your dog. Now, in late 2026, the research has had time to accumulate. The honest summary is that the picture is mixed and the evidence is still young. The most interesting findings concern the choices people make about when to lean on a machine and when to put it down.
Start with how ordinary the habit has become. One industry analysis of the state of AI quoted a technology executive who said that AI has moved from experimentation to behavioral change. That phrase matters. Experimentation is something you do on purpose. Behavior is what you do without noticing.
The numbers back it up. A survey roundup of consumer use found research and question-answering at the top of the list, with 63 percent of AI users turning to the technology for it, ahead even of writing and editing. Another set of statistics found that 45 percent of consumer users say they use AI to write or edit personal messages. Not reports for work. Personal messages. A separate consumer survey from Menlo Ventures found that 19 percent of American adults use AI to help write emails, and another 18 percent rely on it to manage to-do lists or research topics of interest. Payments-industry reporting described most American adults as having now tried consumer AI tools for routine, everyday tasks, with the conversation shifting from whether people will use them to which ones they will keep using.
Younger people are further along. One compilation of usage data found that around half of Gen Z, defined there as ages 14 to 29, use AI at least weekly, and 22 percent use it daily. The same data noted that their excitement about the technology fell by 14 points in a year. So the most frequent users are growing less enthusiastic even as they use it more. That is what a habit looks like. Nobody needs to be thrilled by a thing they do every day.
At work, the pattern is similar. The 2024 Work Trend Index from Microsoft and LinkedIn reported that 78 percent of AI users bring their own AI tools to work, rising to 85 percent among Gen Z workers. Employees are not waiting for an official rollout. They are quietly handing off the small stuff on their own.
Small stuff is the key phrase, because no single handoff looks like a problem. The trouble, if there is one, lives in the accumulation.
The episode continues in the Compound app.