Thinking with Machines: On Brains, Bots, and Being Kind to the Earth

For the past few weeks, I have been training an AI model at work. The goal is simple: to take over the kind of tasks that drain time and energy – the repetitive ones that quietly pile up in a workday. But somewhere between fine-tuning prompts and running test batches, I began to wonder what this really means. Am I making my work smarter, or am I making myself lazier? Are we still using our brains when machines begin to think for us? And while we’re at it, is all this “smart tech” actually smart for the planet?

These questions have followed me from meeting rooms to midnight reflections, and what I’ve realized is this: AI, like most tools, is neither a savior nor villain. It’s a mirror. It reflects how we choose to use it – thoughtfully or thoughtlessly, responsibly or recklessly.

Are We Still Thinking?

When I first started working with AI, I thought of it as delegation. “You do the boring stuff,” I told the model, “and I’ll handle the thinking.” Except, training AI isn’t exactly a hands-off affair. It forces you to define problems precisely, to anticipate outcomes, to understand not only what you want but why you want it. I found myself reasoning more rigorously, not less.

Paradoxically, automation made me more reflective. I had to think about workflows, about efficiency, about the human touchpoints I didn’t want to lose. In that sense, AI didn’t dull my mind – it redirected it. It made me a manager of logic rather than a performer of routine.

Still, there is a quiet danger in that redirection. Once a machine starts handling the small stuff, it’s tempting to disengage. To skim over the output instead of really reading it. To approve without thinking. Slowly, you start outsourcing not just tasks, but thought. And that’s the risk we rarely talk about – not that machines will out think us, but that we’ll stop bothering to think at all.

So I’ve made it a small rule for myself: never stop reviewing. I still do things manually sometimes, just to stay fluent in the work. I let the AI assist me, but I never let it replace the satisfaction of figuring something out for myself. It’s not about distrust — it’s about keeping my cognitive muscles from quietly wasting away.

The Green Question

The more I read about AI, the more I realize it doesn’t run on magic – it runs on energy. Every model trained, every dataset processed, every query generated carries an environmental cost. The servers hum, the cooling systems run and somewhere out there, electricity gets burned to make the thinking happen.

This isn’t to say AI is inherently harmful. In fact, in the right hands, it can make processes leaner, cut down on waste and optimize systems far better than a human could. But efficiency is a double-edged sword. We gain speed, but we often forget to count the miles we’ve traveled to get it.

So, while I build this model to make my own workflow smoother, I also try to keep an eye on its footprint. I choose smaller, lighter models where I can. I reuse trained data instead of starting from scratch. It’s a small demonstration of awareness, proving that the digital world isn’t always eco-friendly.

Perhaps that’s the irony of our age: we’re building intelligence that spans clouds, yet we forget those clouds sit on earth.

The Case for (Careful) Automation

Despite all the caution, I’m still firmly in favor of AI. Not because it’s perfect, but because it’s powerful – when used with care. If it can take the administrative weight off my shoulders so I can think more deeply, create more freely, or simply rest a little more, then that feels like progress.

But I think we owe it to ourselves to use it consciously. To ask not just, “Can AI do this?” but “Should it?” Because some things – like empathy, moral judgment and imagination – aren’t problems to be optimized. They’re experiences to be lived.

I see AI now as a collaborator rather than a competitor. It doesn’t steal my thinking; it stretches it. It encourages me to create systems aligned with my values like clarity, efficiency and responsibility, rather than being overwhelmed by the monotony. But that partnership only works if I stay alert, engaged and human.

Thinking Smarter, Not Smaller

So yes, I’m training a machine. However, I believe it’s also teaching me to be more intentional in my work, to use my energy more consciously and to recognize the hidden expenses associated with convenience.

The question, then, isn’t whether AI will make us stop thinking. It’s whether we’ll use this new freedom to think better.

It’s worth it if I can utilize the time AI gives me to study, be creative and think. Otherwise, I’ve just created a quicker path to thoughtlessness.

For now, I’ll keep training my model – not just to perform well, but to remind me that every click, every output, every byte of “intelligence” still depends on something profoundly human: the choice to use our minds and to care for the world that sustains them.