10 Things Most People Should Ignore About AI (And What Actually Matters)
We've somehow convinced ourselves that everyone needs a PhD-level understanding of artificial intelligence to survive the future.
You don't.
Most people need the equivalent of a driver's licence, not a mechanical engineering degree. You don't need to understand how the combustion engine works to drive to work. You need to know the road rules, how to read the conditions, and when to pull over.
The problem is that the AI conversation, particularly online, is dominated by people who have a vested interest in making it feel impossibly complex, urgently threatening, or magically transformative. And in all that noise, the things that actually matter to real people living real lives get completely buried.
So here's my honest take. Ten things you can safely tune out, and what deserves your attention instead.
1. Ignore: "AGI is coming next year"
Every few months, someone confidently predicts the imminent arrival of superintelligence. The timelines shift constantly. The certainty never does.
The reality? Nobody knows. Not the researchers. Not the CEOs. Not the people making YouTube thumbnails with glowing robot eyes and dramatic music. The history of AI is littered with predictions that didn't land, and a few that arrived much faster than anyone expected. The honest answer is that the future of AI is genuinely uncertain, and anyone telling you otherwise is either selling something or confusing confidence with knowledge.
Here's what I know: worrying about AGI is one of the least productive things you can do with your time right now.
What matters: Learning how to work with AI today. If AGI arrives, you'll adapt, just as humans have adapted to every previous technological shift. If it doesn't, you've wasted years in a state of low-grade existential dread when you could have been building something. The people who will navigate whatever comes next are the ones who are curious, adaptable, and already in motion.
2. Ignore: Which model is #1 this week
Claude. ChatGPT. Gemini. Grok. The next one. The one after that.
The leaderboard changes constantly. Seriously, what was considered the best model three months ago may have already been surpassed twice over by the time you read this. And yet people spend enormous energy tracking these rankings, debating which model is marginally better at coding or slightly more accurate on some obscure benchmark.
Here's what the data actually shows: the productivity difference between a beginner and an advanced user of any given AI tool is almost always larger than the difference between the models themselves. A skilled communicator using an older model will consistently outperform someone who doesn't know how to work with AI using the newest one.
What matters: Learning how to ask better questions. How to give context. How to iterate. How to push back when the answer isn't quite right. A thoughtful, well-constructed prompt beats a model upgrade most days of the week. Invest in your skills, not in keeping up with the leaderboard.
3. Ignore: AI benchmark scores
Models are regularly tested on standardised exams, coding challenges, mathematical reasoning tasks, and obscure academic benchmarks. These scores get reported breathlessly as if they tell you something meaningful about your life.
They mostly don't.
Interesting? Absolutely, if you're a researcher or deeply invested in the technical trajectory of the field. But relevant to helping you plan a holiday, draft a difficult email, organise your finances, think through a business decision, or brainstorm ideas for your next project? Not really.
Benchmarks measure what benchmarks measure. They don't measure whether AI will make your Tuesday easier.
What matters: Whether AI can solve your actual problem. The only benchmark that counts for most people is: did this help me? Did it save me time? Did it give me a perspective I hadn't considered? Start there.
4. Ignore: AI doom and AI utopia
One side says we're all doomed, mass unemployment, existential risk, the end of human meaning. The other says we'll all retire at 35 while AI does everything and we spend our days painting watercolours and going on long walks.
Both are usually selling something. A book. A course. A subscription. A worldview.
The future is almost always messier, slower, and more human than either extreme predicts. The doom narrative requires you to believe that humans will simply stop adapting, which contradicts everything we know about history. The utopia narrative requires you to believe that technology will solve problems that are fundamentally about human nature, power, and distribution, which also contradicts everything we know about history.
What matters: Staying adaptable. Not optimistic or pessimistic, adaptable. Adaptability has outperformed prediction throughout every major technological shift in human history. The people who navigated the industrial revolution, the digital revolution, and every disruption in between weren't the ones who predicted it most accurately. They were the ones who kept moving.
5. Ignore: The idea that AI is "cheating"
People once said calculators were cheating. Then spellcheck. Then Google. Then GPS. Now AI.
There is a long and honourable tradition of humans deciding that the latest tool is somehow a threat to authentic human effort, and then quietly adopting it anyway once it becomes obvious that it just works. The calculator didn't make mathematicians obsolete. It freed them from arithmetic so they could do actual mathematics.
Tools don't remove thinking. They change where thinking happens. The question was never "should I use this tool?" The question has always been "am I using this tool well, and do I still understand what I'm doing?"
What matters: Knowing when to use AI and when to think for yourself. The skill isn't in avoiding AI, it's in developing the judgment to know which problems benefit from AI assistance and which ones require you to sit with the discomfort of figuring it out yourself. That judgment is genuinely valuable. It's also genuinely human.
6. Ignore: Every new AI app
Thousands of AI-powered tools launch every year. Most disappear within six months. Most solve problems you don't actually have, or solve problems you have in ways that are marginally better than what you're already doing.
The shiny new tool trap is real, and it's expensive, not just financially, but in terms of time, attention, and the cognitive overhead of constantly switching systems. Every new tool you adopt requires learning, integration, and the mental load of managing one more thing.
What matters: Having a small, trusted toolkit that you actually use. One AI assistant used thoughtfully every day is worth more than fifty tools bookmarked and forgotten. Find what works for your actual workflow, go deep on it, and resist the pull of novelty for its own sake. Depth beats breadth almost every time.
7. Ignore: The fear that AI knows everything
AI is genuinely impressive in its breadth of knowledge. It has been trained on an extraordinary amount of human-generated text, and it can speak with apparent confidence across an enormous range of subjects.
It is not omniscient. It predicts likely answers, it does not possess universal truth. It has knowledge cutoffs. It can hallucinate convincingly. It reflects the biases and gaps of the data it was trained on. Sometimes it is brilliant. Sometimes it is confidently, completely wrong. And it often can't tell the difference.
This is not a reason to avoid AI. It's a reason to use it with your critical faculties intact.
What matters: Critical thinking. The most valuable skill in the AI era may not be knowing how to use AI, it may be knowing when not to trust it. Knowing when to verify. Knowing when the answer feels too neat, too confident, or too convenient. The people who will use AI most effectively are the ones who bring genuine intellectual rigour to the outputs they receive.
8. Ignore: "Prompt engineering will be the job of the future"
This was one of the earliest and most confidently stated AI myths, and it has aged poorly.
Prompting is useful, genuinely useful. Learning to give AI clear context, specific instructions, and useful constraints will make you more effective. But it is becoming easier every month as models get better at understanding natural language, inferring intent, and filling in gaps. The idea that "prompt engineer" would become a dominant profession was always a bit like saying "search engine query specialist" would be the job of the 2000s.
What matters: Judgment. Knowing what problem is worth solving. Knowing which question to ask before you ask AI to help you answer it. Knowing how to evaluate the output you receive. These are fundamentally human capacities, and they are becoming more valuable, not less, as AI gets better at the execution layer.
9. Ignore: The obsession with replacing humans
Most people imagine AI as a replacement machine, feeding jobs into one end and spitting unemployment out the other. It makes for compelling headlines. It's also a significant oversimplification.
What's actually happening in most industries is more nuanced: AI is changing the composition of jobs, shifting which tasks require human attention, and creating new roles that didn't exist before. ATMs didn't eliminate banking, they changed what bank tellers did, and the number of bank tellers actually increased for years after ATMs were introduced because branches became cheaper to run. Spreadsheets didn't eliminate accountants. Email didn't eliminate communication, it transformed it.
History suggests that technology tends to reshape work before it eliminates it, and that the reshaping creates as much as it destroys. That doesn't mean the transition is painless or equitable, it often isn't. But the binary "replaced or not replaced" frame misses most of what's actually happening.
What matters: Learning to work alongside new tools with genuine skill and intentionality. Not just using AI, but understanding how it changes your role, what it frees you to do more of, and where your distinctly human contribution becomes more valuable as a result.
10. Ignore: The belief that AI is the story
This is the biggest one. And it's the one I feel most strongly about.
AI isn't the story.
Humans are.
The technology is fascinating, genuinely, deeply fascinating. I find it endlessly interesting and I've built my work around helping people navigate it. But the technology is the context, not the content. The real questions, the ones that will determine what kind of future we actually build, are not technical questions.
They are human questions.
What kind of lives do we want to live? What kind of work feels meaningful? What relationships are worth protecting and investing in? What should remain deeply, irreducibly human, not because AI can't do it, but because the doing of it is part of what makes us who we are?
These questions existed before AI. They'll exist long after today's models are obsolete. And the people who are spending all their energy tracking the technology layer are, in many cases, avoiding the harder and more important work of sitting with those questions.
What matters: You. Your values. Your judgment. Your capacity to stay curious and grounded and human in a world that is changing faster than most of us can comfortably process.
The Smart Take
Most people are paying attention to the wrong layer.
They're watching the technology layer, the benchmarks, the model releases, the funding rounds, the predictions. And the technology layer is genuinely interesting. I'm not saying ignore it entirely.
But the people who are navigating this era most effectively are paying attention to the human layer. They're asking what they want to use AI for. They're developing judgment, not just skills. They're staying curious without becoming anxious. They're building systems that serve their lives rather than lives that serve their systems.
AI is getting smarter every month. The real challenge, and the real opportunity, is making sure we do too.
Because the future won't belong to the people who know the most about AI.
It will belong to the people who know the most about people.