The Mahazine newsletter #1

Disney invests 1.5 billion in Epic games, OpenAI launched Sora, Forbes enters the Sandbox metaverse, Skyn Japan uses AI in new commercial, Tasmania fun AI ad, and more...

What happened in the creative AI world in February Week 3&4?

Welcome to the mahazine, a bimonthly newsletter focusing on the latest news in the AI creative world, inspiring visuals and experimentations, trending AI experiences and marketing campaigns, AI tools and tutorials.

📣 Big news

👉 Disney invests 1.5 Billion in Epic games, aiming to create a gaming and entertainement metaverse experience across Disney, Fornite, Marvel, Pixar, Avatar and more… “Players, gamers and fans will be able to create their own stories and experiences, express their fandom in a distinctly Disney way, and share content with each other in ways that they love.” Disney and Epic Games already have engaged hundreds of millions of players through Fortnite content integrations, season collaborations, in-game activations, and live events, including the Marvel Nexus War with Galactus, which drew more than 15.3 million concurrent players.

👉 Roblox removes language barrier with a new real-time AI chat translations, enabling people who speak different languages to communicate seamlessly with one another in their immersive 3D experiences. These translations are displayed in real time, with latency of approximately 100 milliseconds, so the translation happening behind the scenes is nearly invisible. Using AI to automate real-time translations in text chat removes language barriers and brings more people together, no matter where they live in the world.

👉 OpenAI launched Sora, a new AI model to create realistic and imaginative scenes from text instructions. “Sora is able to generate complex scenes with multiple characters, specific types of motion, and accurate details of the subject and background. The model understands not only what the user has asked for in the prompt, but also how those things exist in the physical world.”

👉 Midjourney will ban Trump and Biden image generation in an effort to avoid generations of deepfake political images during the US presidential election.

👉 Forbes Launches Permanent Presence In The Sandbox Metaverse. With its official establishment of permanent land within The Sandbox metaverse, Forbes will offer a variety of interactive experiences, workshops, and events. These initiatives are crafted to bring together minds from various sectors, facilitating meaningful conversations and networking opportunities in a vibrant, immersive environment.

🦄 Mahage explorations

These past 2 weeks have been fun in topics experimentations, we played with for atmospheres and texture, the comeback of the 60’s aesthetic, and wood sculpted toys. Enjoy the pictures of the week and tell us what you think in the comments.

👀 Creative picks

Most creative AI campaigns & ads

[Exhibition] Refik Anadol: Echoes of the Earth The installation wraps the gallery walls in AI generated images inspired by data of flora, fungi and fauna from over 16 rainforest locations globally, which was collected using technologies such as LiDAR and photogrammetry. At the crossroads of visual art, science, and technology. It is the first installation in a growing body of work that is created employing The Large Nature Model, the world’s first open source generative AI model dedicated to nature.

[TV commercial] Japanese Beverage company Ito En, Ltd. A character generated by artificial intelligence in a TV commercial for a green tea drink. The commercial shows an old woman holding the product saying, “I’ll start now for my future self,” and then changes into a young woman. “We decided to use the AI-generated character after thinking about how we could show someone’s appearance in the present and 30 years later”

[Valentine’s day campaign] Skyn Japan: Human softnessCreated by the Japanese studio UltraSuperNew for Valentine’s day, Skin Japan released a video showing a love conversation between 2 AIs. Based on several experiments and hours of research, the team found out that AIs actually do fall in love if we let them talk to each other long enough, with the right prompt. Starting with a casual conversation about the weather, the topic eventually develops into, personal preferences and opinions. They even talk about visiting a coffee shop together and dining together at restaurant one recommended.

[Marketing campaign] TasmanAITo promote Tasmania, Tourism Tasmania and BMF agency imagined a creative campaign introducing TasmanAi, an image generator that uses Tasmanian artists - not algorithms - to turn prompts into real works of art, in the most inspiring place on Earth. People are encouraged to submit their most creative prompt, by mail, and if the artists like it, they'll get sent the real art piece.

[Marketing campaign] ETRO NOWHERE: OUT OF TIME, IN ANOTHER SPACEThe new ETRO SS24 advertising campaign stems from an encounter between imagination and Artificial Intelligence. The inspiration behind the SS24 collection takes the idea of the Nowhere to a higher level of expression, a place of imagination only. Creative Director Marco De Vincenzo converses with the digital artist and prompt designer Silvia Badalotti to create a series of scenes and paintings.

🌞 Curated tweets

🦖 Visual experimentation: Dinosaurs

The creative challenge of this edition is *Dinosaurs*, here is what we are proud of:

Skeleton study: I used Midjourney to generate an image of a dinosaur within a museum. In Photoshop, I removed the dinosaur's body to reveal its skeletal structure intact in the same position. The side-by-side composition highlights how AI and digital tools can be used together to creatively convey scientific concepts through mixed media. This preliminary experiment demonstrated potential for merging art and education. With refinement, similar techniques may help non-experts better understand topics like paleontology, anatomy or even deconstruct paintings layers and techniques.

Fashion study: Have you ever wondered why dinosaur-themed clothing patterns are not more prevalent in stores? Me too. It seems dinosaurs may have untapped potential for bringing vibrant color and drama to runways and closets worldwide. As pop culture keeps prehistoric creatures in the spotlight, their vibrant forms could (must!) take center stage in fashion. I mean, seriously, wouldn’t you wear one of this?

🌊 Technical experimentation: Chaos in Midjourney

What is the impact of different Midjourney parameters on my prompts? In these technical experiments we will test multiple output parameters from different AI tools, in this edition we will focus on Chaos parameter in Midjourney.

What is Chaos? We can read in the Midjourney documentation this description: “The chaos parameter influences how varied the initial image grids are. High chaos values will produce more unusual and unexpected results and compositions. Lower chaos values have more reliable, repeatable results.” 

Let’s try to see how it really impact our prompt.

The prompt:

At dusk from cliffs, a motion camera captures a robed woman against fading light, hands summoning dancing flames from fingertips twisting upon the sea breeze. Below, crashing waves build as sentinels overlook sparks from upwards palms. Creative lighting depicts solitary fire-weaving through sharp detail in dusk's dimming glow.

No Chaos effect

Chaos 20

Chaos 30

Chaos 40

Chaos 50

Chaos 60

Chaos 70

Chaos 80

Chaos 90

Chaos 100

Conclusion: From Chaos 0 to Chaos 50, each image begins to differ slightly and start focusing on different keywords in the prompts, but the change is still small. But after 50 and until 100, we can see the true inner creativity of Midjourney, it takes lot of freedom in the prompt interpretation, proposing camera angles, decors, atmospheres and style of image on its own. To go deeper in the experimentation, we need to understand what is the weight of each keyword in the prompt, what Midjourney prioritize, and how to guide it more into the different criteria of our outcome desires. But this will be in a next technical experimentation.

📰 Updates on AI tools

  • RunwayMulti Motion Brush now available in Gen-2, direct specific movements with a simple brush stroke.

  • Stable DiffusionStable Cascade, setting new benchmarks for quality, flexibility, fine-tuning, and efficiency with a focus on further eliminating hardware barriers. Additionally, they are releasing training and inference code that can be found on the Stability GitHub page to allow further customization of the model & its outputs.

  • Stable Diffusion → Announcing Stable Diffusion 3, improved performance in multi-subject prompts, image quality, and spelling abilities.

  • Heygen → Released Realtime Avatars, multi-modal organic interactions in real-time across voice and video applications

🧠 AI tools

  • Loudfame → Video to Anime

  • Luma → 3D AI

  • Glif → Image generator

  • Debunk → Design suite tools to avoid the blank page

  • Elai → Text to Avatar talking video

  • Skybox → Text to virtual 360 spaces

📖 Interesting reads

🔮 Predictions

  1. The Metaverse will be collaborative: Disney merging digital worlds with Fortnite points the way towards a metaverse experience based on brands partnerships. With the rise of web3, virtual fashion, AI-generated worlds, and collaborative environments and assets created by players, the metaverse will be likely a connection between company experiences, rather than a unique experience owned by one company.

  2. Universal communication: Roblox new AI feature gives hope to ease global communication. As we arrive in the metaverse era, a global open social network, being able to communicate with anybody is a strong feature, that will need to human connection like never before.

  3. Parallel universes: Open AI new model’s Sora is not like the others, it simulates an entire worlds with interdependent physical objects to create a video from the scene. These worlds only exists by prompting them, and they have a short lifespan, what will happen when they will be interconnected, explored, infinite and unrecognizable from reality, or even better than reality?

🎟 Upcoming events

🗼 Tokyo, Japan
The Impact of Generative AI/生成AIが企業活動に与えるインパクト
Thursday, March 7 · 5 - 8pm JSTARCH Toranomon Hills Incubation Center

🎡 London, UK
Blockchain, Web3 , AI,Depin and Digital Assets
Sat, 24 Feb 2024 10:00 - 16:00 GMTSt Michaels Church A503

London AI Startup Summit
Mon, 4 Mar 2024 17:30 - 20:30 GMTUCL BaseKX

🏗 Barcelona, Spain
European Metaverse Research Network: A Metaverse for the Good
Tue, 9 Apr 2024 09:00 - Wed, 10 Apr 2024 17:30 CESTCentre de Cultura Contemporània de Barcelona, Carrer de Montalegre

🖥 Decentraland
Decentraland Art Week 2024 | #DCLAW24
From March 26-29, 2024

If you enjoy the Mahazine, please spread the word by sharing it with your colleagues, friends, and fellow creatives. Your support in growing our community is invaluable as it allows us to reach more creative minds and continue delivering high-quality content.

Join the conversation

or to participate.