A simulation of thinking you haven’t done yet is still just noise, no matter how loud you make it.
I have been doing live sound since I was fourteen years old. Boards, monitors, mains, feedback loops, gain staging, the whole thing. I still keep the skill sharp by hosting karaoke two nights a week, which sounds like a strange way to maintain a craft until you understand what karaoke actually is: a room full of people who almost never sing in public, handing you their most vulnerable moment and trusting you to make them sound like they belong up there. My shows have a reputation for good sound. Non-singers come and stay the whole night just to watch. That does not happen by accident.
The reason it sounds good is simple, and it has nothing to do with gear. I care that when somebody picks up a microphone at one of my shows, they feel like they sound better than they ever thought they could. I am not there to amplify myself. I am there to amplify everyone else.
That distinction is the whole thing.
A Good Amplifier Does Not Editorialize
Amplification is a double-edged sword that most people do not fully respect until they are on the wrong side of it. A good amplifier does not editorialize. It does not add quality that was not already in the signal. It takes what you give it and makes it bigger. Feed it something clean, something with real signal, real presence, real tone, and the result is something that fills a room and makes people feel something. Feed it something muddy, something with noise baked into the source, something that was marginal to begin with, and you get the same thing you started with, just louder and harder to ignore.
Almost everybody can hear when a signal is bad. They may not be able to name what is wrong. They may not know whether it is the EQ, the gain, the room, the mic placement, or the singer. But they feel it. Something is off. Something sounds hollow, or harsh, or muddy, or thin. And critically, something that might have passed unnoticed at conversational volume becomes genuinely unpleasant when you run it through a system and push it into a room.
This is what I am watching happen with AI right now.
AI is an amplifier. A remarkably powerful one. And a lot of practitioners are discovering that they can use it to produce volume they were never able to produce before. More output. Faster. At a scale that used to require entire teams. The problem is that amplification does not care whether your signal is clean. It just makes it bigger.
The Signal Problem
The villain here is not heavy AI use. It is not people who are still learning, because everyone is still learning. The villain is a specific behavior: using AI as a substitute for the thinking you have not done, while presenting the output as if you had. Mistaking the appearance of depth for depth itself, and then amplifying that mistake into a room full of people who are trusting you to know the difference.
I have spent a long time thinking about what I would call the signal problem, which is the gap between what a person actually knows and what their AI-assisted output implies they know. The gap is not always obvious. The output can be polished. The language can be precise. The structure can look like a framework. But when you push on it, when you ask a question that requires the thinking to be load-bearing, you find out very quickly whether there was a data warehouse behind the dashboard.
Time Does Not Automatically Become Craft
The same dynamic plays out at a karaoke show. There are hosts who have been doing this for years who still sound rough. Not because they have not put in time, but because time alone does not build skill. It just builds comfort with your current level. The people who get genuinely good are the ones who stay curious, who listen critically to their own output, who can hear the difference between what they are producing and what they are trying to produce, and who care enough about the gap to close it.
Pride is the enemy of that process. And there is a particular kind of pride that comes from having done a thing for a long time. It is the pride that says years of effort should translate automatically into quality, that the work of showing up is sufficient proof of the quality of what you are delivering. It is not. Half of everybody is below average, including people who have been trying for years. The uncomfortable math does not change just because the effort was genuine.
Taste is the piece people skip over. You can learn technique. You can read about frameworks. You can study what good looks like and build a working vocabulary for it. But taste, the ability to hear the difference between clean and muddy before you push it into the room, that is built slowly, through exposure and failure and genuine caring about the result. You do not know what it sounds like until you hear it fixed. And once you hear it fixed, you cannot unhear it.
Amplification For Whom?
Ideas deserve to be amplified. A genuinely good idea can come from anywhere, and the goal of any decent practitioner should be to find those ideas and give them room. But the idea has to actually exist before you turn up the gain. A good sound engineer cannot save a singer who is not singing. They can flatter the performance, hide some of the rough edges, make the room feel warmer than it is. But there is a floor. Below a certain signal quality, no amount of craft behind the board changes what people hear.
The reason my shows sound good is not because I have better gear or more years behind a board. It is because I am not there for myself. Every decision I make is in service of the person holding the microphone, who trusted me with something vulnerable. That orientation, outward instead of inward, changes every choice downstream of it.
That is what I am looking for in AI-augmented work. Not just a clean signal, but a clear sense of who the amplification is actually for. Because if the answer is you, if the goal is to appear more capable, more credible, more prolific than the underlying work warrants, the audience will feel it eventually. They always do. They may not be able to name it. But something will sound off.
Build enough craft, care, and taste that amplification serves people instead of ego. Then turn it up.