Man, it feels like every other week there’s a new AI model promising to change everything. But is the hype real? I just watched a video from an AI professional who got early access to GPT-5, and I have to say, I’m genuinely blown away by the tests he ran.
This is way beyond just writing a poem or summarizing an article. This innovator was having it build entire interactive applications from scratch.
✨ Building Interactive Worlds from a Prompt
The creator didn’t just ask it to write code; he had it build entire interactive simulations that you could play with right in the browser.
- 📌 The Rubik’s Cube: First up, he prompted it to build a fully interactive, 3D Rubik’s Cube. Not only did it create a beautiful, realistic cube, but it could handle scrambling and solving for different sizes, all the way up to a massive 20x20x20, which it solved step-by-step right on screen!
- 📌 3D Game of Life: Then, he asked for Conway’s Game of Life with a twist: visualized on 3D shapes like a sphere, a donut (torus), and even a Mobius strip. The result was absolutely stunning.
- 📌 Complex Physics: The YouTuber also had it build a complex double pendulum simulation with accurate physics and a cloth simulation with tearable constraints and wind effects. It handled all the complex math and visualization flawlessly.
⚙️ Full App Clones… In One Go
This is where my jaw really dropped. The person who shared it asked GPT-5 to build clones of major software with just a single prompt.
- ✅ Excel Clone: He literally just said, “create me a clone of Excel,” and it spit out a working version with multiple sheets, cell formatting, and even support for formulas like =B8+C8. That’s just wild.
- ✅ Word Clone: He did the same thing for Microsoft Word! It generated a front-end with text formatting, lists, undo/redo, and even image insertion.
- ✅ 3D Lego Builder: It also created a photorealistic 3D Lego builder where you could place and rotate different colored bricks. The visual quality was seriously impressive.
👁️ Next-Level Multimodality
GPT-5’s ability to understand images is on another level. This is not just describing what’s in a picture.
- 💡 Geolocation: The creator uploaded a vacation photo from his trip and just asked, “Where is this?” GPT-5 nailed the location, identifying the county and even the specific road he was on in Northern California.
- 💡 Recreating from an Image: He gave it a screenshot of a physics simulation he saw on Twitter and simply said, “Recreate this.” And it did, building a nearly identical interactive version from that single image.
This is just a fraction of what this industry pro packed into this video. He also tested its speed, reasoning abilities (it gives surprisingly solid and safe life advice), and its skill at creating gorgeous front-end dashboards and web pages.
I think this shows we’re entering a new era of what’s possible with AI. For the full deep-dive and to see all these insane demos for yourself, you have to watch the original video from the creator!