🧠 THIS WEEK IN CREATIVE AI Issue #50

Here are the 5 big stories from these past few weeks that every creative working in Generative AI should know about:

🎬 1. Runway Aleph: Edit Video With Words

Runway just soft-launched Aleph, a new model that makes in-footage video editing feel like magic. Think: change time of day, adjust pacing, or even shift camera angles without re-generating the entire scene. It’s modular, non-destructive, and signals a future of video that adapts to your story, not the other way around.

💡 Why it matters: Video just became liquid. (borrowing a phrase from Matthieu Lorrain ) - Creative teams can now experiment with narrative like building blocks.

🌀 2. Midjourney Video End Frame + Loop

Midjourney Video now supports End Frame Referencing and Perfect Loops. This means you can reference the final frame of one image to seamlessly transition into the next — a game-changer for anyone creating carousels, scroll-stoppers, or looping motion art.

🎨 Try this: Build seamless image sequences that flow like motion design… but made entirely in Midjourney.

As shown in the Caption Video for this newsletter, the legendary Rory Flynn has already been putting the new Midjourney End Frame through its paces - original post.

🌐 3. Pika Gets a Web UI Overhaul

Pika just dropped a redesigned web interface that feels faster, sleeker, and much more intuitive for multi-shot editing. Think DaVinci Resolve-lite, but AI-native. Alongside it: better prompt anchoring, smoother transitions, and easier version comparison.

🛠️ Why care: The barrier between casual video ideation and polished production just got thinner, like a lot thinner.

🦾 4. Leonardo Phoenix 1.5: Better Control, Better Faces

Leonardo.Ai 's Phoenix model got a quality-of-life update with v1.5. It’s now significantly better at rendering consistent faces, dynamic lighting, and artistic styles - especially with its new “Coherence” slider. Great for character design and scene recreation.

🔥 Try it: Create comic-style scenes with facial consistency.

🗣️ 5. ElevenLabs Multilingual Voice AI Expansion

ElevenLabs now supports over 30 languages with emotion-aware TTS and a stronger focus on voice consistency across cloned characters. Dub content into multiple languages with tone and cadence intact - no robotic aftertaste.

🌍 Use case: Globalise your ad, explainer, or trailer in hours, not weeks.

TL;DR: Creative AI platforms are moving fast - from video editing with words, to image loops that feel like motion, to multilingual voiceovers on tap.

Now, at edition 50 of this newsletter, we can confidently predict that the future’s not just generative. It’s fluid, remixable, and soon to be ready to scale!

✌️ - The Seedily Team -