AI

OpenAI just released Sora 2 —their flagship video and audio generation model. I’ve said many times that Sora sucks at video generation, but I’ve changed my mind with this new version. Sora 2’s videos are far more physically accurate and realistic (see here and here). It still produces occasional flaws, but the success rate is […]

Over the past few years, ChatGPT updates have left many guides outdated. Models and prompting techniques have evolvedOpenAI added and removed some ChatGPT features GPT-5 was introduced, changing how ChatGPT works In this guide, I’ll show a clear path to go from beginner to pro in ChatGPT, updated for 2025. Each stage of this path

The iPhone 17 has just been released, and I’m already putting it to work to get the most out of ChatGPT. After a year of using the ChatGPT app on my previous iPhone, I’ve identified some features that have boosted my productivity. Here are three that have changed how I use ChatGPT on my phone.

Made with ChatGPT I’ve been using ChatGPT since the day it was released. Back then, there were no fancy features, model picker, or alternative AI tools to choose from. Things have changed in 2025 and I’d like to share with you some things I’ve learned so far: A couple of times per year, you’ll see

I used to struggle to create and edit images with ChatGPT-5. It sometimes didn’t understand my instructions, generated text with typos, and more. Luckily, after understanding the changes made after the GPT-5 release, I was able to have ChatGPT generate and edit the images the way I wanted. Here’s what I did.