The Cat Was Never on the Counter

Four months after Brooklyn disappeared, somebody is still sending his owner pictures.

That is where I want to begin. Not with the clever software. Not with the men who call this a trust problem while they sell another trust solution. With the messages arriving after the cat is gone.

A woman gets a photograph. There is Brooklyn, apparently safe, sitting on a kitchen counter while somebody holds him. For one clean second, the room changes shape. The bad thing has ended. The cat is alive. He can come home.

Then comes the request for money.

Then the close look at the picture. The cat is posed just as he was in the missing-pet photo. The syrup bottle in the background has writing on it that has gone soft and drunk. The microwave has been rearranged by a machine that knows the look of a kitchen but has never had to wipe one down after a bad night.

No cat. No kitchen. No rescue.

Just a small manufactured miracle with its hand out.

I have lost animals. Not in the precise way this woman lost Brooklyn, maybe, but enough to know the mind becomes a cheap detective when an animal is missing. You walk streets you have already walked. You look under cars. Every orange shape at the edge of your vision gets a second look. You hear a noise behind a fence and your heart, that old fool, gets up and runs before the rest of you has decided whether it is worth the trouble.

A lost pet is a hole in the ordinary day. You put food in a bowl out of habit. You see the place where they slept. You do not stop listening.

So naturally somebody found a way to turn that listening into a scam.

There is no technical mystery in it worth admiring. Give a machine a photo of a missing cat. Ask for the cat on a counter, by a doorstep, being fed, being held. The machine does not need Brooklyn. It does not need a counter. It needs the pattern of a frightened person wanting both those things to be true.

Then some bastard sends the result to the person who is frightened.

This is the part of the artificial-intelligence circus I keep coming back to. The machine itself has no cruelty. A hammer does not hate your thumb. But somebody picks it up. Somebody sees a woman’s missing-pet post and thinks: there is a soft place. I can put a hook in it.

The new machine just makes the hook look like her cat.

The old internet had its dishonest corners. You learned not to wire money to princes. You learned that a stranger with a miracle investment was usually trying to finance his own lunch. But even the con needed a little labor. A liar had to invent details. He had to send bad spelling from an email address that looked like it had been assembled during an electrical storm.

Now he can send a photograph.

Photographs used to be lousy evidence, of course. People posed. Cameras lied by omission. Newspapers cropped things. Your uncle showed slides of a vacation designed to prove he had not spent the whole week drunk by a pool. But a photograph had a body behind it. A camera was in a place. Light bounced off something. Somebody could at least ask what happened before and after.

The fake cat picture has none of that. It is a sentence with no speaker, a footprint with no animal, a hand reaching through the mail slot wearing a glove made of pixels.

And it does more damage than one stolen payment.

The people documenting real cruelty to animals are already getting it in the neck. Welfare groups publish footage from farms, laboratories, circuses—the ugly places where animals become units and the units become somebody’s business model. The footage is hard to watch. That has always been part of its power. Now the first answer underneath can be: AI.

Maybe it is. Maybe it is not.

The doubt is cheap. That is why it spreads.

A person who does not want to look at a row of calves in hutches does not need to prove the film is false. He only needs to introduce enough fog. Was it generated? Was it staged? Is there a watermark? Can you show me the original file, the editing history, the camera, the date, the weather, the birth certificate of the person holding the lens?

Meanwhile the animal stays in the cage, which is a hell of a convenient outcome for everyone except the animal.

The same flood buries the good stuff. Real rescues lose attention to machine-made polar bears being saved from impossible floods by heroes with six fingers and perfect cheekbones. Actual shelters compete with accounts pumping out synthetic puppy misery for clicks. Donations are pulled toward whatever makes the thumb stop moving, and the machine can make a hundred drowning animals before lunch without getting wet.

The people selling us all this smooth poison will say labels are coming. Invisible tags. Detection tools. Better provenance. Fine. I am for anything that helps a grieving woman see the trap before she sends money to it.

But do not ask me to confuse a label with a cure.

The scammer does not care about a label. He will use another tool, strip a tag, change a pixel, open a new account. The platforms will call it a difficult enforcement problem because every problem gets difficult when the profitable part of it is allowed to happen at industrial speed.

And the rest of us will become little forensic clerks in our own lives. Before we believe the lost dog was found, before we donate to a wildlife rescue, before we let a photograph move us, we will squint at the background text and examine the paws. We will become suspicious of tears because a company decided every feeling was raw material waiting to be rendered.

That is the real theft. Not merely money. The first rush of relief. The instinct to believe a cat is safe. The ordinary human reflex that sees a creature in trouble and wants to help.

Brooklyn is still missing.

Somewhere his owner may still be checking doorsteps, corners, alleys, the same old places where a real cat might be. The fake pictures keep arriving, each one pretending to close the wound and opening it again.

The machine can make the cat sit anywhere.

It cannot bring him home.

Source: AI Slop Is Ruining Cute Animals on the Internet