I Changed the Way I Write AI Image Prompts, and This Is What Happened

I see this happen all the time.

You sign up for a new AI image generator.

You’ve seen the amazing, photorealistic stuff other people are posting.

So you sit down, excited to try it.

You type in: “A delicious burger.”

And you wait.

Then… you get a cartoon.

Or worse, you get something that looks like plastic.

The cheese looks like yellow play-doh. The lighting is flat.

It looks like I hired a photographer and a food stylist.

But it took me about 45 seconds.

That’s the "secret."

You don't need a $5,000 camera, perfect lighting, or a photography degree to shoot a perfect photo.

Give these 5 AI principles a try next time you create an image.

You’ll see the difference immediately.

It just looks fake.

I used to have the same problem.

I thought maybe the "pros" were using a different version of the software than I was.

But they weren't.

I found out they were just talking to the AI differently.

It turns out, AI is a lot like a contractor.

If you give vague instructions, you get a vague result.

But if you give specific instructions, you get exactly what you want.

So I stopped trying to "guess" the right words.

Instead, I built a simple checklist.

I call it C.L.A.S.P.

I use it every single time I write a prompt.

It’s working very well.

My images went from looking like "AI Art" to looking like high-end photographs.

And I’m not doing anything technical.

I’m just making sure I hit these 5 specific points.

Here is how it works:

C = Camera

This is the one everyone forgets.

If you don’t tell the AI which camera lens to use, it guesses.

And it usually guesses wrong.

So I tell it exactly what I want.

If I’m shooting food, I tell it to use a "Macro Lens" or an "85mm Lens."

You need to force it to simulate real glass.

• For Portraits: Specify an "85mm lens" (flattering, zoomed in).

• For Landscapes: Specify a "16mm wide-angle lens."

• For Details (Food/Products): Specify "Macro lens" or "100mm."

Pro Tip: Add phrases like "Depth of field" or "Bokeh" to blur the background. This instantly makes an image look expensive.

L = Lighting

Bad lighting makes food look cold.

Great lighting makes it look hot.

I never just say "lighting."

I say "Golden Hour" if I want it warm.

Or I say "Cinematic Lighting" if I want it dramatic.

This is what triggers the appetite.

Another example is "Volumetrics" (hazy/steamy), "Softbox lighting" (studio clean).

A = Atmosphere

Context is everything.

A burger floating in white space is boring.

A burger on a "rustic wooden table" in a "busy kitchen" tells a story.

I always give the AI a background; I describe the air around it.

For example: "Dust particles dancing in the light," "Crowded rainy street," "Rustic wooden table," "Cyberpunk neon haze."

S = Subject

This is the easy part. It's the thing you are taking a picture of.

But here is the trick:

Don't just name the food. Describe the texture.

Instead of "Steak," try: "Charred ribeye steak with glistening fat and coarse salt crystals."

Instead of "Woman," try: "Portrait of a woman with freckled skin texture and wind-swept hair."

The AI needs to know how it feels.

P = Presentation

This is the final polish.

I tell the AI what style I want.

• "8k resolution."

• "Food Magazine Style."

• "Hyper-realistic."

• "National Geographic style."

• "High-speed photography."

Here is the difference it makes.

Most people write a prompt like this:

“A photo of a bowl of ramen.”

And they get a clip-art bowl of soup.

But when I use C.L.A.S.P., I write this:

“Shot on Sony A7R IV with a 100mm macro lens (Camera). Soft cinematic lighting highlighting the steam (Lighting). Dark rustic wood table background (Atmosphere). Spicy Tonkotsu ramen with a soft-boiled egg and jammy yolk (Subject). 8k resolution, hyper-realistic (Presentation).”

The result?

You get a steaming, mouth-watering image where you can actually see the texture of the pork.