AI should be better

AI should probably be better than it is by now.

Not to sound ungrateful for groundbreaking technology, but by its nature, AI learns by being used. The more it’s used, the smarter it should get, and the rate of improvement should be exponential. Multiplied. Compounding.

I was reviewing old LinkedIn posts for content inspiration and spotted something I wrote three years ago, mocking an AI image where the eyes were melting and the hands had extra fingers. Genuinely terrifying stuff.

And while looking into that fresh hell of an image I thought, “AI images are pretty damn good now.” Deepfakes are scarily accurate. The hands mostly behave, and the eyes don’t look like they’re dissolving.

But it’s funny, back in 2023 we were told marketers, graphic designers, illustrators would be dead by now.

And yet, AI still can’t make art or content that humans actually find compelling.

That’s the bit nobody seems to want to admit. The technical leap has been enormous. The taste gap hasn’t closed at all. I’m not seeing AI-generated essays I want to read, AI-generated music I want to listen to, or AI-generated videos I’d choose over a human-made one. The technology that was supposed to flatten creators in eighteen months has, three years in, only made better stock photos.

So, if you’re someone trying to build a portfolio career in this environment, that should give you some hope. Not because AI won’t keep improving (it will), but because the thing you bring to the table, the thing that makes your work yours, is still the part the machines haven’t cracked.

Here’s the thing though. If you buy into the narrative that you’re about to be replaced, you will be replaced. That’s not a tech problem, that’s a you problem. The portfolio careerists who’ll thrive over the next decade are the ones who figure out who they are, learn how to express that on the page and in their work, and stack skills that compound on each other.

Be the kind of person other people want to buy things from. Build the products and services only you could build.

Do that and you don’t need to panic. Don’t do that and no amount of “AI-proofing your skillset” listicles will save you.

So don’t give in to nihilism. Don’t give in to what’s-the-point-ism. The technology is going to keep marching forward whether you spiral about it or not, so you may as well keep moving forward.