Alright, so you wanna know about that “did steve harvey box” thing I was messing around with? Buckle up, it’s a wild ride of late nights, questionable decisions, and a whole lotta Googling.
It all started with me seeing some meme online, right? Some joke about Steve Harvey looking like a boxer or something. And my brain, being the way it is, went “I wonder if I could make that real?” Not, like, actually make Steve Harvey box (that’s insane), but could I generate an image of him in a boxing ring using AI?
First thing I did, obviously, was Google. I searched for stuff like “AI image generation”, “text-to-image models”, and “how to make Steve Harvey a boxer”. The results were overwhelming. So many different tools and platforms, it was kinda dizzying.
I ended up settling on this one online AI image generator – can’t remember the exact name, honestly, there are so many. Figured it’d be the easiest way to dip my toes in without having to download a bunch of stuff or write any code. Signed up for a free trial, ’cause I’m cheap like that.
Then came the fun part: trying to get the AI to understand what I wanted. I started with simple prompts like “Steve Harvey boxing” and “Steve Harvey in a boxing ring”. The results were… not great. Lots of weird, distorted faces and backgrounds that looked like they were melting. Think Salvador Dali meets a fever dream.
Okay, so I needed to be more specific. I started adding details. “Steve Harvey boxing ring, wearing boxing gloves, serious expression”. Still kinda wonky, but getting closer. The gloves looked like they were attached to his head in one image, and another one had him fighting a teddy bear, but hey, progress, right?
This is where it got time-consuming. I spent hours tweaking the prompts, trying different combinations of words and phrases. “Steve Harvey, boxing match, dramatic lighting, sweat on his brow”. “Steve Harvey, victorious boxer, crowd cheering, championship belt”. Each time, I’d run the prompt, see the results, and then tweak it again. It was a total trial-and-error process. The AI seemed to have its own weird interpretation of everything.
Some of the funniest fails? One image had him fighting a giant cheeseburger. Another had him wearing a tutu. I swear, the AI was just messing with me at that point.
After a bunch of tries, I finally got something that looked vaguely like what I was going for. It wasn’t perfect, mind you. The lighting was a little off, and his face was still a bit… AI-generated-looking. But it was close enough. I’d say about 70% there.
To try and fix it, I used another free online tool to upscale the image and sharpen it a little. Helped a bit, but still not perfect. But after hours, I was willing to call it a win and I was calling it a day.
So, “did Steve Harvey box?” No, of course not. But did I generate a kinda-sorta-believable image of him boxing using AI? Yeah, I did. Was it worth the time and effort? Maybe not. Was it a fun distraction? Absolutely.
The biggest takeaway? AI image generation is cool, but it’s still kinda janky. You really have to play around with the prompts to get anything decent. And even then, the results can be pretty unpredictable. But hey, that’s part of the fun, right?
- Googled AI image generation.
- Signed up for a free trial of an online AI generator.
- Experimented with different prompts.
- Tweaked prompts based on the results.
- Upscaled and sharpened the final image.
- Decided it was “good enough”.
Final Thoughts
Would I do it again? Probably. There’s something strangely addictive about trying to bend the will of an AI. Plus, the fails are hilarious. Maybe next time I’ll try to make Gordon Ramsay a ballerina.
