Opinions vs AI: Why Your Brain Still Matters

I have opinions. Lots of them. And after decades in technology and manufacturing, I’ve earned every single one through late nights debugging systems, failed implementations, and hard-won insights that only come from doing the work.

Lately, I’ve noticed something that concerns me: colleagues increasingly using AI as their primary—and sometimes only—source for complex topics. Don’t get me wrong, AI is incredible. But there’s a difference between using it as an accelerator and using it as a replacement for thinking.

The Shortcut Trap

Last month, I was deep into research on implementing group-wide KPIs. I’d been methodically working through several books, taking notes, connecting concepts, and building my understanding of the nuances. When I mentioned this to a colleague, they immediately punched the title of one of my key references into an LLM.

The AI summary they got back was good—accurate, well-structured, comprehensive. But it missed the subtleties and nuances that made that particular book so valuable for what I was trying to accomplish. Had we relied solely on that summary, I’m confident we wouldn’t have been as successful or effective in our implementation.

This isn’t a condemnation of AI. It’s a warning about how we use it.

AI as Accelerator, Not Replacement

Here’s how I actually use AI in my work: as an accelerator, not a substitute for expertise.

For basic research: AI helps me quickly identify relevant frameworks, methodologies, or case studies to explore further. It’s like having a research assistant who can point me toward promising directions.

For validation: When I’ve formed opinions based on experience and research, I’ll use AI to stress-test my thinking, find counterarguments, or identify blind spots I might have missed.

For synthesis: After doing the hard work of understanding complex topics, AI can help me articulate or organize my thoughts more clearly.

Notice what’s missing? AI isn’t forming my opinions or doing my thinking. I’ve already done the hard work of building expertise and establishing my knowledge base.

The Expertise Problem

I’m particularly worried about younger software developers who might be tempted to skip the difficult learning phase. There’s no shortcut to understanding complex concepts or building the judgment that comes from working through hard problems.

But it’s not just the junior end of the spectrum that concerns me. We’re also watching really great developers and engineers retire with decades of irreplaceable experience—the kind of deep, nuanced knowledge that no generic LLM is going to help you with. That institutional knowledge, those hard-won insights about why certain approaches work and others fail spectacularly, that understanding of subtle interdependencies—it’s walking out the door.

When you let AI do your thinking, you’re not just missing information—you’re missing the mental models, the intuitive understanding, and the hard-earned wisdom that comes from wrestling with difficult concepts yourself. You’re outsourcing the very process that builds expertise.

Your brain is like a muscle—it gets stronger with use and weaker without it. The more you exercise your critical thinking, pattern recognition, and problem-solving abilities, the better equipped you become to collaborate effectively with AI. When you have strong foundational knowledge and well-developed analytical skills, AI becomes a powerful multiplier of your capabilities. Without that foundation, you’re just getting sophisticated (and potentially wrong) answers to questions you don’t fully understand.

The Art of Doing the Work

The best business leaders, technologists, and professionals I know share a common trait: they’ve put in the time to truly understand their domains. They’ve read the books, made the mistakes, debugged the systems, and built the mental frameworks that let them see patterns others miss.

This isn’t about being anti-AI or nostalgic for the “good old days.” It’s about recognizing that AI works best when it amplifies human intelligence, not when it replaces it.

The Moral of the Story

Do the work and challenge yourself. Don’t be afraid of using AI, but use it as an accelerator not as a replacement for the spark of innovation and creativity.

My Challenge to You

• Read the full book, don’t just get the summary
• Work through the hard problems yourself before asking for help
• Form your own opinions based on experience and research
• Use AI to enhance and validate your thinking, not to do it for you

The future belongs to people who can effectively combine human insight with AI capabilities. But that combination only works if you bring real insight to the table.

Do the work. Challenge yourself. Build expertise that matters.

Your future self—and your organization—will thank you for it.

Now go do something awesome.

What’s your experience with balancing AI assistance and developing genuine expertise? I’d love to hear your thoughts at [email protected].