Brain Training in the Age of AI: Why Staying Sharp Matters More Than Ever

There's a quiet trade happening every time you reach for an AI assistant. It writes the email, so you don't compose the sentence. It summarizes the article, so you don't read it closely. It remembers the fact, so you don't. Each individual trade feels trivial — even smart. But repeated thousands of times, those small handoffs add up to something worth paying attention to: we are doing less of our own thinking than at any point in modern history.

That isn't an argument against AI. AI is genuinely one of the most useful tools ever built, and pretending otherwise helps no one. It's an argument for being deliberate. If a tool quietly takes over the reps that used to keep your mind in shape, the responsible move isn't to abandon the tool — it's to keep training the muscle on purpose.

The cost of letting something else do the thinking

Psychologists call it cognitive offloading: using external tools to reduce the mental effort a task would otherwise demand. Writing a reminder instead of memorizing it is cognitive offloading. So is using a calculator, a GPS, or a large language model. Offloading is often rational — your memory is finite and your attention is precious. But it has measurable effects on what your brain bothers to retain.

The classic demonstration came well before ChatGPT. In a 2011 study published in Science, researchers Betsy Sparrow, Jenny Liu and Daniel Wegner showed that when people expect to be able to look information up later, they remember the information itself less well — but remember where to find it better. The popular name for this is the "Google effect," and it captured something true: when an external memory is reliable and always available, the brain treats it as part of the system and stops doing the work itself.

Generative AI raises the stakes, because it doesn't just store facts — it does the reasoning. Several 2025 studies have started to map what that means:

• A team at Microsoft Research and Carnegie Mellon surveyed 319 knowledge workers about how they use generative AI at work. They found that higher confidence in the AI was associated with less critical thinking, while higher confidence in one's own skill was associated with more. When people trusted the tool, they shifted from doing the work to merely verifying it — and sometimes skipped the verifying, too.
• A study by researcher Michael Gerlich, published in the journal Societies, found a negative correlation between frequent AI-tool use and critical-thinking scores across roughly 666 participants, with cognitive offloading appearing to mediate the relationship — especially among younger users.
• A widely discussed preprint from the MIT Media Lab, "Your Brain on ChatGPT," used EEG to compare people writing essays with an LLM, with a search engine, and with nothing at all. The LLM group showed the lowest brain connectivity and struggled to quote from essays they had just "written." It's a small, not-yet-peer-reviewed study and shouldn't be over-read — but it points in the same direction as the survey work.

None of this says AI makes you permanently dumber. It says something more specific and more actionable: the skills you stop using get rusty. Which is, conveniently, the whole point of the name.

The goal isn't to use AI less. It's to stay sharp enough to use it well.

The skills worth protecting

If you're going to deliberately train anything in the AI era, train the abilities that make you a better director of AI rather than a passive recipient of it. Four stand out.

Critical thinking. AI is fluent, confident, and sometimes confidently wrong. The ability to spot a weak argument, notice a missing assumption, or sense when an answer is too clean is now a competitive advantage. You can't outsource the judgment of whether the output is any good. For more on this, see our guide on how to improve your critical thinking.

Clear communication. When everyone can generate a polished paragraph in seconds, the scarce skill becomes knowing what's actually worth saying — and being able to say it crisply, out loud, in a meeting, without a tab open. Articulation is thinking made audible.

Mental math and estimation. Not because you'll never use a calculator, but because a fast internal sense of numbers is how you catch errors, sanity-check a model's output, and make decisions in real time. If you've lost it, our piece on mental math tricks is a good place to restart.

Focus. Every one of the above depends on being able to hold attention on something hard for more than a few seconds. Deep focus is the substrate; everything else is built on top of it.

What actually keeps you sharp

Here's the honest part, because it matters for trust: brain training is not magic, and anyone promising that puzzles will raise your IQ is overselling. The research is clear that you reliably get better at the specific things you practice, and far less clear that those gains transfer to unrelated abilities. (We dig into that nuance in does brain training actually work?.)

But that nuance points to a practical strategy rather than away from one:

1. Train the real skill, not a proxy for it. If you want to think and speak more clearly, practice thinking and speaking — not abstract tile-matching. Specificity is a feature, not a bug.
2. Favor consistency over intensity. A few focused minutes every day beats an occasional marathon. Habits compound; cramming doesn't.
3. Get feedback. Practice without feedback just rehearses your current level. Feedback is what turns repetition into improvement.
4. Keep doing the reps AI would happily take from you. Draft the hard sentence yourself first. Estimate before you calculate. Recall before you search. Then use AI to go further, not to skip the rung.

How Rusty fits

This is exactly what we built Rusty to do. It's a five-minute daily "circuit" of short, game-like reps across mental math, critical thinking, focus, memory and communication — the skills above, trained directly rather than by proxy. And it leans into the irony of the moment: Rusty puts AI on your side of the table. When you answer a speaking prompt out loud, it transcribes your response and gives you an instant Clarity Score — grading structure, reasoning and expression, counting filler words, and telling you the one thing to fix next time.

The point isn't to compete with AI. It's to make sure that when you use it, you're still the one doing the thinking — the editor, not the autocomplete.

AI will keep getting better at thinking for us. The most valuable thing you can do in response is refuse to let that be the only thinking that happens. Spend a few minutes a day keeping your own mind in shape, and you get the best of both worlds: a powerful tool, and the judgment to wield it well.

Want to start today? Download Rusty free on the App Store and run your first circuit in five minutes.