TL;DRI don't know what I believe about AI anymore.Some days I'm excited. Some days I'm scared. Some days I'm just tired of the shouting. And honestly? That's okay.We don't need to pick a team. The useful thing isn't being "for AI" or "against AI." The useful thing is staying honest about how complicated this feels—and letting myself sit with the discomfort long enough to understand what's actually at stake.Also, heavy metal got there before LinkedIn did.\There's something exhausting about almost every conversation I have about AI lately.Someone tells me AI will make us faster.Someone else tells me it will make us lazy.Someone says it will democratize knowledge.Someone else says it will flatten creativity, steal jobs, and put power into fewer hands.And then the room splits.One side talks like AI is salvation. The other talks like it's the end credits of civilization. Both speak with the confidence of prophets and the patience of a comment section.I don't want to take a side here.Not because I don't care. Exactly the opposite. I care enough to think this deserves more than a team jersey.AI is not a football club. It's not metal vs pop. It's not tabs vs spaces, although I fully understand why someone would try to make it that way.AI is a tool, a market, a cultural shift, a risk surface, a workplace accelerant, a legal headache, a creative companion, a surveillance temptation, and a mirror. Sometimes it's useful. Sometimes it's nonsense with a nice interface. Sometimes it's both in the same five minutes.So maybe the better question isn't:"Are you pro-AI or anti-AI?"Maybe the better question is:"What am I choosing when I use it, ignore it, regulate it, trust it, fear it, or build on top of it?"That's a much more honest conversation.And, weirdly enough, heavy metal has been teaching me this lesson for decades.\The AI Debate Is Really a Debate About ChoicesMost arguments about AI aren't really about AI.They're about control.Who gets more of it? Who loses it? Who profits? Who becomes replaceable? Who gets amplified? Who gets copied? Who decides what "good enough" means?That's why I react so differently depending on the day.A developer using AI to speed up boring boilerplate may feel relief.A junior developer trying to get hired may feel panic.A teacher may see a cheating machine.A student may see a private tutor.A writer may see a collaborator.Another writer may see a machine trained on the unpaid labor of people like them.A founder may see leverage.An employee may see headcount math.All of these reactions can be rational at the same time.This is the part I keep forgetting. Different people aren't always disagreeing because one side is stupid. They're often standing in different places, with different risks, different incentives, and different things to lose.That doesn't mean every opinion is equally informed. Some takes are lazy. Some are fear-based. Some are hype wearing a leather jacket. But disagreement itself isn't a bug. It's a signal.When a technology becomes big enough to touch work, culture, money, education, privacy, and identity, disagreement is the cost of admission.\Metal Song #1: Metallica — "The Unforgiven"Metallica's "The Unforgiven" isn't about artificial intelligence, obviously. But it's very much about pressure, identity, and what happens when someone is shaped by forces around them before they get to choose for themselves.That makes it a useful song for the AI conversation.One of the quiet fears I have around AI isn't simply that people will use it. It's that people will be pushed into using it before they have time to understand what they're giving up.Use the tool or fall behind.Automate the task or lose budget.Generate more content or disappear from the feed.Adopt the system or become "inefficient."This is where the AI debate becomes personal for me. A choice stops feeling like a choice when the environment punishes you for choosing differently.The lesson here is simple: before we ask people to adapt, we should ask what kind of adaptation we're demanding from them.Are we giving people agency?Or are we just renaming pressure as innovation?\The Lazy Version of the Debate Helps NobodyThere's a lazy pro-AI argument:"Stop complaining. This is progress. Adapt or die."There's also a lazy anti-AI argument:"Everything about this is theft, slop, and doom."Both are emotionally satisfying. Neither is enough.The "adapt or die" argument ignores power. Not everyone has the same time, money, safety, or bargaining position to adapt.The "everything is doom" argument ignores usefulness. Many people are already using AI in practical, non-apocalyptic ways: summarizing documents, debugging code, translating ideas, exploring concepts, or getting unstuck.I need a better middle.Not a weak middle. Not a "both sides have a point" shrug. A stronger middle.A place where I can say:AI can be useful, and it can be harmful.AI can expand access, and it can centralize power.AI can help people create, and it can flood the world with low-effort garbage.AI can save time, and it can raise expectations until nobody actually gets that time back.AI can be impressive, and it can still be wrong.That's not a contradiction. That's adulthood.\Metal Song #2: Black Sabbath — "The Mob Rules"Black Sabbath's "The Mob Rules" is a perfect warning for any moment when crowds start thinking in slogans.AI discourse has a mob problem.Not always an angry mob. Sometimes it's an investor mob. Sometimes it's a workplace productivity mob. Sometimes it's a social media outrage mob. Sometimes it's a futurist mob with a pitch deck.The problem with mobs isn't that they're always wrong. The problem is that they move too fast for nuance.And AI requires nuance.Should a doctor use AI? Maybe, but not without accountability.Should a student use AI? Maybe, but not in ways that destroy learning.Should companies use AI for hiring? Maybe, but now we need to talk about bias, transparency, appeals, and legal responsibility.Should artists use AI? Maybe, but now we need to talk about consent, attribution, and compensation.Should governments use AI? Maybe, but please do not make me trust a black-box system with someone's benefits, border status, or prison sentence.The mob wants a chant.Reality wants a checklist.And I'm tired of being asked to choose a chant.\A Better Checklist for Thinking About AIBefore picking a side, I try to ask myself:Who benefits if this AI system works?Who pays the cost if it fails?Is there a human appeal process?What data trained it?What data does it collect now?Can people opt out without punishment?Does it replace judgment or support judgment?Are we measuring quality, or just speed?What happens to beginners if AI takes over beginner tasks?Would I still want this system if I were the person with the least power in the room?That last question is the big one.A lot of AI looks great from the manager dashboard.It may look different from the employee inbox.It may look different from my own inbox.\Metal Song #3: Judas Priest — "You've Got Another Thing Comin'"Judas Priest's "You've Got Another Thing Comin'" is pure defiance. It's the sound of refusing to let someone else write your future for you.That energy matters to me in the AI age.Because whether I'm excited about AI or worried about it, passivity is the worst option.If I love AI, I shouldn't just cheer for it. I need to learn how it works. Learn where it fails. Learn what it costs. Learn who's missing from the room where it gets deployed.If I hate AI, I shouldn't just reject it from a distance. Test it. Understand it. Know the thing I'm criticizing well enough that my criticism cannot be dismissed as fear.If I'm undecided, good. I should stay undecided a little longer. But stay actively undecided. Read. Experiment. Ask people outside my bubble what they're experiencing.The future isn't only built by the people who say yes.It's also shaped by the people who ask, "Wait, why?"Bonus Track: Iron Maiden — "The Wicker Man"Iron Maiden's "The Wicker Man" has the feeling of standing at the edge of something big, loud, and unavoidable.That's also how the AI moment feels to me.There's a door opening. There's noise behind it. Some people want to run through. Some people want to lock it. Some people are still trying to understand who built the door, who owns the building, and why the exit signs are written in legalese.The point isn't that everyone must walk through at the same speed.The point is that pretending the door isn't there won't help.And I can't pretend anymore.\Be Open, But Not Empty-Headed"Be open-minded" is sometimes used as a trick.People say it when they really mean, "Please stop questioning my business model."That's not what I mean here.Being open-minded about AI doesn't mean accepting every tool, every claim, every roadmap, or every CEO prophecy.It means holding multiple possibilities in my head without rushing to simplify them.It means being able to say:"This is useful, but the business model worries me.""This is creative, but the training data question matters.""This saves time, but it may also deskill people.""This is not good enough today, but it might be tomorrow.""This is impressive, but I still need evidence."That's the kind of thinking I need more of.Not blind optimism.Not fashionable pessimism.Just serious curiosity.And maybe a little more honesty about how scared I am sometimes.\The Real ChoiceThe real choice isn't AI or no AI.That ship has sailed, caught fire, raised funding, and launched an API.The real choice is how I think while the technology spreads.Do I become a fan?Do I become an enemy?Do I become part of the mob?Or do I become a careful user, builder, critic, teacher, regulator, artist, citizen?The heavy metal answer would probably be: don't kneel, don't sleepwalk, and don't let the crowd choose your mind for you.That sounds about right.AI is complicated.Good.Complicated things are where thinking is still required.And I'm still here, trying to think.