Andre Assalino

The latest from the Visualise Immersive R&D labs - we've taken live underground train data from Transport for London and overlaid it on their maps using #augmentedreality. You can hold up your phone to a TFL map and see exactly where the trains are - in realtime. It's just a bit of fun but a great way to see how augmented reality can bring data to life and make the unseen far more tangible. Big thanks to Callum Fowler, Jack Østergaard-Churchill, Anna Firbank and Juliet Alliban for bringing this one to life! #datavisualisation #augmentedreality #londonunderground Ping us if you want to have a play!

111

12 Comments

Would you guess this is AI-generated or shot for real? This entire Sun Cola stadium sequence I generated in 10 mins using AI workflows with only minimal graphic overlays added in After Effects. What’s becoming scary good is the consistency: - Natural crowd reactions in high quality - Stable product appearance without morphing - Realistic and natural camera movement - Facial emotions that feel surprisingly real - All sound effects generated from the same video prompt A few months ago, shots like this would completely fall apart. Now, AI video is getting dangerously close to real commercial production quality — and it’s becoming easier to make bold AI production decisions without the same fear of morphing, broken frames, or unstable results that held productions back before. The tools are evolving fast, but the workflow and creative direction still make the biggest difference. Open to collaborations for cinematic AI ads and branded content.

42

7 Comments

AI-Powered Animation Experiment: There’s been so much buzz about AI lately, with everyone promising the world. So, I decided to dive in and see for myself what AI can really do for us right now. This video is the result of that experiment: a fully animated piece created entirely with AI tools. Every element—script, lyrics, music, visuals, and animation—was AI-generated. (Well, except for the editing—that was my handiwork in iMovie 😉) Here’s the AI lineup + quick review: • ChatGPT – script and lyrics: Super easy to collaborate with, fast and surprisingly creative. • Suno – music composition: Blown away by the quality. Catchy, polished, and fast. • Midjourney – visual art: Still one of the best tools to create images with prompts. The latest model update (v7) has improved drastically the character consistency. • Hailuo AI – animation: Promising tool! Still rough around the edges, like transition animations of body parts like arms, eyes and hands… but very impressive for early-stage animation. • iMovie – editing: Manual, but effective. This was a creative challenge, a learning experience, and honestly, a lot of fun. Would love to hear what you think—and if you’ve tested any of these tools yourself! #ai #artificialintelligence #animation #chatgpt #suno #midjourney #hailuo #creativity #futureofwork

22

1 Comment

£12,000 in prizes. 150+ confirmed participants. One route to Cannes. Big Screen Hack is bringing filmmakers, AI artists, engineers and designers together for a 48-hour creative-AI sprint exploring the future of filmmaking. Teams will compete for £12,000 in cash, credits and partner prizes from ElevenLabs, Panalux, Runware, MiniMax, Vercel Workflow, imgn AI, Adi Sideman, Misfit Productions, RecDek and Videostack.ai. But this is not just about prize money 🤑 It’s about putting the newest creative-AI tools into the hands of people who understand story, lighting, performance, editing, sound, motion and workflow — and pairing them with the builders who can turn experiments into the next generation of filmmaking tools. Two tracks: 🎬 Make a film — create a 48-hour AI-native short 🛠️ Build a product — create tools for the film industry The best films will be shortlisted for a screening at Cinéum Cannes during the Cannes Film Festival. 🤯 The best products will be shown to the industry that needs them. Free to enter. London + remote. 48 hours. Two tracks. One route to Cannes. https://luma.com/bsuuql7j

125

7 Comments

"One-person studios will keep doing great work for simpler formats: social media, short ads, quick content, but higher-end AI films will need teams again." The future of AI filmmaking isn't just better models, it's better collaboration. Love seeing professionals like Luis Eduardo Amaral already sharing thoughts on Preview publicly, despite the product being unannounced and in Private Alpha. Hoping to launch publicly soon and finally give professionals the tools they deserve. Demand through referrals over the last 2 weeks has been incredible, we barely have time to sleep from onboarding new studios.

34

2 Comments

Tom ELPHNT has very kindly included Flowfal in the latest additions to his Max For Live Curated Collection🤘 https://lnkd.in/er7z9Ysd So we thought we’d make a demo showing Flowfal controlling one of the ELPHNT devices, in this case the ANLG preset collection. We chose the Arp Mod preset for this demo. A set of simple chords are fed into the Arp Mod device which then produces some lovely arpeggios. Using one of our Flowfal digital wands (it works in just the same way using a phone or smartwatch) we can use the Flowfal Map device to map the “roll” of the wand to three of the Arp Mod parameters: Decay, Pluck and Colour. This allows you to really play around with the texture of the sounds being produced. Note that in this case we have set it up to make the Pluck parameter move in the opposite direction to the Decay and Colour parameters - but you can map anything you like to the roll movement. We have also mapped the “pitch” movement of the wand to the number of arpeggiator steps - so tipping the wand up will start moving notes up one or more octaves, depending on how far up we tip the wand. Lastly, we have mapped acceleration from the wand to the arpeggiator rate. When the wand is moved around a bit the rate is increased giving the effect of fast arpeggio runs. With Flowfal you can map any kind of movement from a phone, smartwatch or wand to any mappable parameter in Live - so the options are limitless. #abletonlive #maxforlive #musicproduction #musicproducer #flowfal

11

So this week has been really interesting with all the developments from Runway and Luma AI offering similar technologies. I’m honestly blown away by these advances, and if I were in any of these creative partner programs for either company, I’d be playing with these platforms right now, but I’ll have to wait. Another thing happening is that Google has decreased their prices on Veo 3, which suggests they’re not as competitive as they used to be, this was part of a marketing plan, or perhaps they want to keep their current customers glued to the platform until open source catches up. Recently, Wan 2.2 came out, and from the examples I’ve seen, it’s outperforming both Kling 2.1 and Veo 3 in three separate text to video tests. So what does this mean for AI video? It probably means that none of the closedsource video models are safe against opensource rivals like Wan 2.2, and it’s only a matter of time until Veo 3, Kling and others are overtaken by low cost opensource video models, and we’re all forced to start learning ComfyUI, lol. I think this will change how people work with AI video in the future. I’m still interested in how the opensource LLM market is doing, so I check in every now and then. I used to check every day, but since advancements have slowed a bit, I’m tuning in only every few weeks. There’s a new LLM called GLM that’s outperforming many of the closedsource paid models like ChatGPT and Gemini. It’s incredible how quickly open source is catching up. This week was the first time I noticed both AI video models and LLMs becoming much more competitive against closedsource offerings. It probably won’t be long until there’s almost no control over advanced AI models, and we might see a new conceptual leap beyond deep learning into the next frontier.

68

18 Comments

🚀 Just wrapped up a 10-second 3D animation using Cinema 4D + Redshift, built around the idea of a result-publishing process. The concept? Think of a smart system that takes in files (like images, PDFs, videos, DOCX), scans them through a futuristic tunnel, and finally publishes a grade. From modeling to animation to rendering — every part of this was done from scratch. This project challenged me to think visually about how we process, evaluate, and display data in a clean and creative way. 🛠 Tools: Cinema 4D + Redshift 📌 Role: Everything – concept, design, execution 📺 Final output: A cinematic-grade animation with a bold finish (literally: an A 😄 haha) Would love to hear your thoughts, feedback, suggestions, or even collab ideas are always welcome. #3DAnimation #3DUiAnimation #SaasVideoAniamtion #Cinema4D #Redshift #MotionDesign #VisualStorytelling #CreativeDirection

1

2 Comments

Consistency remains the ultimate challenge in AI video generation. Moving from abstract concepts to stable, cinematic sequences requires a deep understanding of framing, precise camera movements, and tool limitations. This project, titled "What It Costs," focuses on that exact challenge—bridging the gap between generative tools and tight visual continuity. The goal isn't just to create a striking standalone frame, but to build a stable, narrative-driven visual language that holds up to a cinematic standard. For the creators and directors in my network: What workflows or platforms are you currently finding the most success with for maintaining character and asset stability? #AIVideo #GenerativeVideo #CreativeDirection #CharacterConsistency #AIWorkflow #VisualStorytelling #CinematicAI

2

Omni Editing just dropped, and it’s a game-changer Leonardo.Ai Omni Editing, powered by FLUX.1 Kontext and GPT-Image-1 is here! Every image has its inline prompt bar, allowing you to relight scenes, shuffle objects, refine layouts, and rewrite text in seconds:

118

6 Comments

Saudi Arabia doesn't need a filter. It needs a frame. Last month I produced this cinematic film entirely with AI no on-ground crew, no traditional production budget just a vision, AI tools, and hours of craft. No dialogue. Only the visual power, mesmerising landscapes & cinematic camera movement. The result: visual storytelling that competes with traditional productions, at a fraction of the cost and time. This is the future of content marketing for organisations like Red Sea Global and Saudi Tourism Authority | الهيئة السعودية للسياحة and every DMO or travel brand ready to embrace AI-driven production. Waiting for your comments. #AIVideoProduction #DestinationMarketing #SaudiArabia #CinematicAI

1

2 Comments

Adobe has released a major AI update for Premiere Pro, and I’ve taken some time to break down the features and what they could mean for video creators across Bristol and the South West. From automated rough cuts to improved audio clean-up, these changes have the potential to streamline a lot of everyday editing tasks. You can read the full blog here: 👉 https://lnkd.in/e_QKCfeG

13

🎥 Luma AI strikes back: introducing “Modify with Instructions” for Dream Machine Just write what you want, and watch your video change. Literally. Luma’s new Modify with Instructions lets you edit any video using a single prompt: ✨ “Change the red dress to a blue suit” ✨ “Make it golden hour” ✨ “Replace the coffee cup with a wine glass” 📌 What you can do: 🔸 Remove or replace objects from your scene 🔸 Add rain, snow, fog or change seasons 🔸 Restyle outfits, makeup, or even entire environments 🔸 Insert virtual products or props 🔸 Redesign sets, textures, or color schemes 🧪 Currently rolling out to Dream Machine subscribers, and more features are coming soon. 🔔 Follow me for the last AI updates! #LumaAI #DreamMachine #AIvideoediting #AInews #GenerativeAI

73

5 Comments

Why does almost every AI animation still look like Pixar? Not because the tools are limited. Mostly because that is the visual language people already know. When most people first start experimenting with AI, they default to polished 3D animation because it feels familiar, safe, and instantly recognisable. But as a result, a huge amount of AI content ends up looking almost identical - boring. There are so many other animation styles worth exploring. Try using this prompt on ChatGPT instead (prompt in comments). This approach pushes images into a more surreal and eccentric version of reality. Elongated proportions, distorted atmosphere, imperfect motion, strange tension. Less polished. More personality. I originally found this prompt on Threads, but I could never track down the original creator. If anyone knows who made it, please tag them below. #generativeai #genai #aivideo #aifilmmaking #seedance

184

24 Comments

Best examples of how the promise of AI is great but doesn’t always deliver Don’t get me wrong, I love AI. But tools are only ever as good as the craftsperson using them. The best results come when skilled people use AI as an extension of their craft, not as a replacement for it.

2

I’m trying to get back into the habit of attending interesting events and talks, so last night I went to a Runway meetup organised by Cinesite. It featured four speakers who’re working with Runway video (and other AI tools): first Jon Biggs showed how Monks is using Runway as part of a vast agentic media creation workflow, then artists Ryan Phillips (Uncanny Harry), Roxanne Ducharme (TrashCanRoxanne), and Matt Burn (Oranguerillatan) showed off examples of their work and talked about their creative processes. Each of these artists has come to AI later in their careers (TV directing, animation, and CG/VFX, respectively) so it was super-interesting to see how their experience informs and elevates their generated shorts. Host Andy McNamara then talked through some of the latest features in Runway, with insight into how a 30-year-old VFX and animation studio is adapting to rapid change. There was plenty for the AI-positive and AI-curious to think about. I’d go along to another of these in a heartbeat.

25

8 Comments

Bad Design, Good Direction: A New Way of Working with AI After three years building and working with AI, I created a practical case study showing what's actually possible when you start directing AI. We are living through the biggest shift in human history... an age that requires us to learn how to tame machines and control them. I hope you enjoy it. For more content like this, join my free newsletter! Link in comments.

209

53 Comments

There are a few AI tools that let you 'reimagine' videos, and the latest is from Luma AI using their Ray 2 video model. The feature is called Modify Video and is effectively a video-to-video tool that lets you change the style of the original source clip. Over the past couple of weeks I've seen lots of examples, from people changing themselves into barbarians with lightsabers, to gunslinging in the old west. It's all very fun. But is there a way of using this in the real world? I think so. Combine a tool like this with on-brand visuals, and you could start creating proofs of concept, animatics and mood films. What this won't do is replace designers, illustrators and animators. Will that change? Possibly as the tech improves, but for now this is a tool that helps and enhances open-minded creatives. This example features our fab marketing exec Roxanne Jane Relusco and is just a quick technical experiment, but I see the potential. It's also available in Definition AI, our new tool with all the best AI in one place, for your brand.

262

15 Comments