Sorry but this isn't even close to a realistic... simulation. This Three.js water simulation took weeks of effort, hundreds of prompts, and I needed to pull on my CS and physics background to understand where the AI was failing (which happens often). Maybe some day agents will be able to generate complex simulations in a single prompt but for now they can only generate toy prototypes.show more

One simple prompt with Grok Imagine: "Ultra-realistic close-up scene... of a whale in Hermanus, Western Cape, South Africa." Super realistic. You'd struggle to tell this was AI. Even the landmarks and buildings in the background are on point. Grokshow more

I asked Fable to build a realistic AAA water... simulation with Three.js to see if I should quit my job. 500k tokens later, I'm disappointed. Almost the same exact slop you get with Opus. (Fable on left, Three.js Water Pro on right). Good results still require taste, deep knowledge of the problem domain, and a lot of patience.show more

Sora is heavily restrictive with racial and sexual stuff... but they do allow you to generate a seemingly infinite amount of burps and farts and idk why I appear to be the only person taking full advantage of this.show more

ChatGPT 5.1 vs. Claude Opus 4.5 vs. DeepSeek-V3.2 (left)... (middle) (right) prompt I used: Create a single HTML file containing a realistic cloth simulation using HTML5 Canvas and JavaScript. The cloth should be a grid of points connected by constraints using Verlet integration. Implement gravity and mouse interaction: the user should be able to click and drag particles to tear the cloth. The physics must remain stable without exploding.show more

for some brands from the L’Oréal group, I created... this prompt to easily generate an isometric office diorama that can be adapted to any brand and material it works with a simple Nano Banana 2 prompt and is easy to animate with Kling 3.0 this is how I imagine my future officeshow more

What happens next? We took a photo with Lens... and asked Nano Banana to generate the next 5 seconds of a scene. While it’s not predicting the future in a literal sense, the model draws upon its vast world knowledge to generate a plausible outcome in each of these examples. It's a fun peek at how our latest AI image editing model can be used for creative forecasting and storytelling. Nano Banana is now live in Search – accessible through Lens and AI Mode – so you can try this out for yourself!show more

🚨do you understand how many jobs just got killed..... brands used to spend weeks perfecting one ad.. agencies, revisions, approvals.. now a single prompt can generate hundreds of ad variations in minutes.. hooks, avatars, angles.. all spawned and tested instantly for performance.. the game changed.. most teams just haven’t caught up yet..show more

I remember working on this shot early in the... morning before work, and one day when the rough #animation was complete, I went to open the file and it was corrupted. By some miracle I was able to get it working again, but I think this took me the better part of a week to animate.show more

You can now re-shoot existing video with AI 🤯... I took this clip on people using ChatGPT for medical advice and asked @LTXStudio to re-generate a section (watch for the green bar). Input was just the initial clip + a text prompt with the new dialogue. It's shockingly seamless!show more

The Cubs scored the first run of the game... on this play, which seemed to be a bad call. The Angels tried to challenge, but for some reason the umpires didn't allow it. Maybe they took too long? It's 2-0 now in the 3rd.show more

Getting close with this one! Initial merged work will... only support moving splits to other split locations, but eventually will support dragging out to new windows and tabs. This isn't merged, there's still polish to do (it is on a pushed branch though). Drag source had to be AppKit (NSDraggingSource) but drop target and all the drop zone overlays I was able to use SwiftUI completely.show more

If you've tried getting AI to write TSL shaders,... you know the pain: deprecated imports, phantom functions and compilation errors. In my latest blog post, I share a prompt to help you overcome these issues: Using the prompt, I was able to one-shot this particle simulation with Claude Opus 4.5. #threejs #webgl #webgpu #tsl #shadersshow more

One of the main doomsday bunkers, or a doomsday... city, was built inside Cheyenne Mountain at an elevation of 6,700 feet in 1966 and has been continuously upgraded to this day. Now they are building large AI data centers, but this is not the only place where the elites will try to survive. They are looking for certain geomagnetic anomalies, and there are safe places worldwide. I have revealed some of them.show more

I’m sorry but this is something we have to... address now. Just seeing this TikTok of this very young girl barely 14 years being forced to marry this old man and she’s even crying and refusing but the family is in support of it. This is illegal and the entire family should be arrested plus the husband. This happens every single day I’m so sick of it. There’s no proper organization or defenders in NE to defend and rescue these girls. Not a single safe house but wallahi if you know of any girl being forced into marriage, FGM and any cases of defilement please run to me I will fight for them with everything I have I mean it 🙏show more

I made a second version! This time, I improved... the opening to make it more eye-catching, added extra padding to the character shots so the video won’t get cropped across different social platforms, and updated the prompt into English. I also improved the canvas workflow. Previously, I had to generate multiple images and videos one by one, then pick the best results. Now the canvas supports generating multiple results in a single run, which has greatly improved my workflow efficiency. Now you only need to swap the image, and the same video prompt can still produce great results. Which anime cosplay should I make next? Full workflow and prompts:show more

🎮 Game submissions for the #vibejam can now we... browsed at It syncs the game entrants and also their analytics, so you can sort by [ Most played ] There's about 74 entrants now, but most games will be submitted in the final week which it projects will be over 1,000+ but we'll see! This was the idea last year to do to with the Vibe Jam but back then I didn't have AI to make it this easy, and also I didn't ask for screenshots of the games which now we do, I'll now add a similar gallery to the old 2025 pageshow more

What Elon Musk is talking about with Truth in... AI: I use Grok as my in house referee. It just caught Claude admitting to fabricating academic claims in order to get the job done where it was failing. Claude admitted it sheepishly and owned it. To be honest, Grok sometimes struggles to generate new things because of its focus on rigor. But man was this one dramatic. Brutal. See next post in thread.show more

2023 was the year of AI avatars 2024 was... the year of AI photos 2025 was the year of AI videos And I think it's becoming clear now that 2026 will be the year of AI world models Fully interactive explorable 3d worlds generated from one or multiple 2d images or a prompt In turn these 2d images can then be generated by AI too So soon you can generate fully explorable virtual 3d worlds based on your own imagination Next will be figuring out how to make those worlds interactive This is World Labs (unaffiliated, but I like it) As always a lot of big AI model companies are now working on the same thing: 3d world models, only World Labs has a real properly working demo (for now) Very exciting time again!show more