Quick heads-up for the AIography crew — Kling has officially released v3 along with the new o3 models, and this is a meaningful moment in the AI video space. Kling 2.6 has already been widely lauded for its quality and feature set, to the point where many higher-end creators I follow are treating it as the current state-of-the-art video model — in some cases surpassing Sora, Veo 3.1, and Runway for real-world creative work. What makes this v3 / o3 release interesting is that it builds directly on what made 2.6 special, rather than reinventing the wheel: - Longer, more coherent shots that hold together narratively - Stronger subject and character consistency across clips - Noticeably improved motion logic (less drift, fewer artifacts) - Tighter native audio sync - Overall behavior that feels more intentional and cinematic Important note: v3 / o3 is currently available only to Kling Ultra plan subscribers, with a broader rollout expected soon. This looks very much like a staged release rather than a limited experiment. Bottom line: Kling continues to separate itself by focusing on creative reliability and control, not just raw generation. If you’re serious about AI-assisted filmmaking, this is a release worth tracking closely. If anyone here already has Ultra access and has hands-on impressions, I’d love to hear what you’re seeing.

Hey everyone! 🎬 The new issue of the AIography newsletter just dropped and it's a big one. THE LEAD: Disney invested $1 billion in OpenAI and licensed 200+ characters for Sora. The mouse is officially in the AI game. This is INCREDIBLY HUGE news for every type of creator. I do my best to break down what it means for all of us making stuff in this space. THREE TUTORIALS THIS WEEK: 1. Jay E from RoboNuggets n8n workflow that pumps out broadcast-quality ads for under $3 (full breakdown of the cost structure—this one's a game changer) 2. Tao Prompts 7 AI video prompt styles that actually work: Timestamp prompting, anchor prompts, cut-scene prompting, and more 3. TechHalla's grid prompting technique for multi-character consistency (actually posted this one here for you guys already!) ALSO: Runway dropped 5 world model announcements. They're not just making video anymore—they're building reality simulators. I explain why this matters. VIDEO OF THE WEEK: @nouryyildiz's "Hollywood Selfie Part 2"—running through classic Hollywood sets with Brando, Eastwood, DiCaprio. 2M views. Made with Nano Banana and Kling. Just pure fun. Oh, and I'm committing to at least one tutorial in every issue from now on. The tools are evolving fast—we need to keep up. ONE MORE THING: Still not subscribed? Get on it! It's free and it's the easiest way to stay current in AI filmmaking without drowning in noise. And if I can ask: if you're getting value from this, share it with one person who needs to see it. A fellow editor, a filmmaker friend, that creative who keeps asking you "what tools should I be learning?" This is the answer. https://aiography.beehiiv.com/subscribe Finally, let me know what you think. I value every one of your opinions good or otherwise. Hey, I was an editor for decades, I'm used to people tearing apart my work! 😆✌🏼

My wifey Simcha-Chaya filled in for me (with some minor trolling at the end from hubby) on the YouTube channel to show off Higgsfield AI's new Cinematic Studio feature! https://youtu.be/KYISSt99vho?si=bSvsyA3ehTxWDA6V

Hey everyone 👋 I’ve been working on something for AI filmmakers and just launched it this weekend. It’s called ILUM — a space to share AI-made films & series, get visibility and (if you want) monetize your work. Not trying to spam — just sharing because many here create amazing stuff.If you want to take a look:✨ www.ilum-stream.com And if anyone wants to be part of the first group of featured creators, happy to chat.

Hey AIographers — big news! OpenAI just dropped Sora 2, their next-gen video + audio generation model, and from the looks of it, it’s a quantum leap. Here’s what’s getting me hyped (and what you’ll want to experiment with): - Handles physical realism and failure states (think: if you miss the hoop, the ball doesn’t teleport to it). - You can drop yourself or others into generated scenes via “cameos” photo-real voice + appearance fidelity. - Dialogue + sound effects are built in, not just visuals. - OpenAI is launching it via a new Sora app (iOS first) and invites via rollout. - They’re taking safety seriously: consent, likeness control, teen limits, moderation, etc. - Free tier initially, with premium / paid tiers later. Imagine scripting a short film in which you insert yourself mid-scene, with fully synced dialogue and scenery that obeys real world physics. That’s what Sora 2 is aiming for. It’s not perfect, it still slips sometimes, but it feels like the jump from text → image to video is happening now. I’ll pull apart its strengths, weaknesses, and what this means for creatives in a breakdown soon. Stay tuned! For the time being, Below are just a few demos from the Sora 2 page. If you want the full scoop, Check it out HERE.