Our brain is not ready for AI (yet)

Our brain isn't built for AI. Maybe it never will be.

We have now access to multiple brains (AI) but still only one to manage all of them. And we end up doing more than one thing poorly instead of one thing well.

Here are some reflections I wrote down while using AI.

The Multitasking Trap

AI tools make it so easy to switch contexts.

One minute I’m generating code, the next I’m asking Cursor to explain a library or querying some new ideas for my new feature.

It feels productive - I’m doing all the things! - until I realize my focus is shattered.

Research backs this up.

Interruptions from AI suggestions can disrupt our flow and actually increase cognitive load by forcing frequent task-switching.

In one study, programmers said Copilot’s pop-ups "messed up" their thought process, and even when they ignored the suggestions, their concentration was already broken​. In short, every context switch has a cost. Our brain needs time to refocus after each little AI detour, and those seconds add up.

I’ve started being more mindful of this. If I’m deep in coding, I try resist the urge to fix every error with Cursor immediately. More multitasking can mean less actual work done!

Source: Multitasking doesn't save time - it wastes time. "The One Thing" by Gary Keller

Too Many Ideas, Too Little Focus

Then there’s the flip side: sometimes AI gives us so many ideas that our brain overheats.

Ever asked ChatGPT for brainstorming and gotten a dozen options - only to feel weirdly paralyzed?

Or spent an hour tweaking an image prompt in MidJourney because maybe the next variation will be perfect? I’ll admit I have. It’s like analysis paralysis on steroids. Our cognitive energy gets drained by evaluating endless AI-generated possibilities, leading to decision fatigue.

Surprisingly, using AI doesn’t always lighten our mental load. It turns out that vetting AI outputs - deciding which code suggestion to accept takes its own mental toll. Each suggestion is another decision, another context switch and it can become exhausting.

What's helped me is setting some boundaries for code generation.

• If something breaks, don't try too hard to fix it. Revert to a previous state, change the model, or adjust the prompt. When you "talk" too long with the model, it always tries to fix its mistakes and creates new ones along the way.

• I almost never iterate on a problem more than three times. This means that I will stop generating and look at the code to see where the problem lies.

I'd rather not constantly hit 'refresh' for more. Limiting the options available to me helps me to avoid being overwhelmed and maintain momentum.

Chasing the AI High

Let’s talk about the dopamine rush we get from these tools. When Cursor magically completes that function or ChatGPT gives us the answer we needed, it feels good – a mini reward.  

Code generators are like a slot machine: keep feeding prompts and you’ll eventually hit the jackpot. Each "win" gives a quick hit of satisfaction, and it’s tempting to seek that feeling again…and again.

Recent research is starting to compare this to addictive behavior. One academic paper noted that ChatGPT's instant, on-demand responses provide instant gratification that can reinforce compulsive use. Essentially, the more quick answers we get, the more we crave them. Over time, we might find ourselves opening ChatGPT reflexively, even for things we could figure out on our own, just because we’ve grown used to that fast reward. Another researcher described it as a dopamine-driven loop – completing tasks with AI boosts our sense of accomplishment, which can push us to take on even more work to get that next dopamine hit​.

I’ve felt this personally: "That solution worked so well, what else can I solve tonight?" Suddenly it’s midnight and I’m knee-deep in a new project.

Understanding this pattern is crucial. It doesn’t mean I ever will ditch AI, but I’m learning to be aware of the urge behind the usage.

Take a short walk, go to gym, what ever brings me down and resets the impulse. The goal is to keep AI as a tool I control, not a compulsion that controls me.

Late Nights and Anxious Mornings

Because these AI assistants are available 24/7, they can sneak into every corner of our day.

I’ve had nights where a random idea hits me, and next thing I know I’m in Deep Research rabbit hole.

It feels productive in the moment, but I pay the price later – my sleep gets disrupted and my mind races with too many ideas.

I think many of us have felt that "brain won’t shut off" syndrome.

Striking a Healthier Balance

So, how can we continue to reap the benefits of these incredible AI tools without overloading our systems?

At the end of the day, awareness (not only attention) is all I need. AI is transforming how we work and create, but it doesn’t change the fact that we’re humans with cognitive limits and psychological needs.

By recognizing the subtle ways AI tools pull at our attention and reward circuits, we can use them more mindfully.

For me, it’s about making these tools serve my goals without undermining my focus, creativity, or well-being. With a bit of discipline (and yes, the occasional digital detox), we can enjoy our AI helpers and still get a good night’s sleep - brains recharged and ready to build the next cool thing, come morning.