👉 Midjourney is transforming into a collaborative platform

Midjourney has launched a new multiplayer collaborative worldbuilding tool named 'Patchwork'. This new standalone platform allows users to collaboratively create and share expansive digital worlds, integrating various elements for immersive storytelling. Patchwork features an infinite canvas with a toolbox that includes options for creating characters, events, factions, and more. Users can generate new worlds by entering text prompts and selecting from various image styles, producing a workspace filled with AI-generated assets and "scraps" for further development. The tool allows for real-time collaboration, where multiple users can interact within the same world.

Future updates will enhance character consistency across images and explore the creation of immersive 3D environments. Midjourney plans to roll out additional features, including multiple model personalization and video models, with a new version, Midjourney V7, set to improve prompt understanding.

👉Pika 2.0 lets you mix your own characters, objects, and scenes

Pika has launched Pika 2.0, a significant update that enhances user control and customization in video creation. This new version allows users to upload and integrate their own characters, objects, and scenes, making it easier to create engaging videos. The platform boasts improved text alignment and motion rendering, addressing previous challenges in generative AI video tools.

The introduction of the “Scene Ingredients” feature allows for granular customization, appealing to influencers and advertisers alike. As major brands such as Balenciaga and Vogue leverage Pika for creative advertising, the platform is set for continued growth. Founder Demi Guo emphasizes Pika's mission to foster creativity and storytelling, paving the way for broader adoption of AI video technologies.

👉 Twelve Labs enables AI search through video content

Twelve Labs is pioneering a video analysis AI capable of understanding and searching through video content in innovative ways. Co-founder Jae Lee emphasizes the limitations of traditional keyword search methods for video, as they fail to capture the actual content of clips. Twelve Labs’ models enable users to search for specific moments, summarize clips, and answer detailed questions about video content. The company has garnered support from major investors, including Nvidia, Samsung, and Intel, and has raised a total of $107.1 million. Their technology supports diverse applications, including ad insertion and content moderation, and aims to address issues like bias in AI training by developing robust ethical benchmarks.

👉 LetzAI launches Sigma Mode for enhanced image generation

LetzAI has unveiled a new feature called Sigma Mode, designed to optimize the image generation process for users seeking faster and more detailed results. Sigma Mode operates at a 720p resolution, delivering images that maintain the high quality LetzAI is known for, while also allowing for intricate details such as realistic skin textures and pixel art. This new mode is approximately 40-60% faster than the default settings and costs only half the credits—making it one of the most affordable high-quality image generation options available, at under $0.02 per image.

Sigma Mode is particularly beneficial for rapid experimentation, achieving photorealistic results, and creating specific aesthetics like VHS-style or 16-bit art. The mode uses a technique known as Lying Sigma Sampling, which minimizes image artifacts and enhances perceived quality.

In addition to the new mode, LetzAI is introducing new image ratios: Sigma Mode supports ratios up to 9:4/4:9, while the Default Mode now extends to 8:3/3:8. Sigma Mode has a maximum resolution of 1440x1440 pixels, but 720p resolutions such as 1280x720 are recommended for optimal results. Both Sigma and Default images can be upscaled directly on the platform.

👉 Yell Group and Alibaba Cloud Partnership

Yell Group, a prominent creative digital company based in Thailand, has entered a landmark partnership with Alibaba Cloud during the Alibaba Cloud Global Partner Summit 2024 in Bali, Indonesia. This collaboration aims to revolutionize Yell Group’s AI-Deate platform, a storyboard AI tool that currently empowers creators in 102 countries. The partnership seeks to meet the rising demand for generative AI solutions in the creative media industry. By leveraging Alibaba’s scalable, reliable cloud-based solutions, Yell Group aims to enhance scalability and drive innovation in the industry.

👉 Brandtech leverages Adobe Firefly for enhanced marketing efficiency

The Brandtech Group has partnered with Adobe to integrate Adobe Firefly Services into its generative AI marketing platform, Pencil Pro. This integration enables seamless interoperability between Adobe tools like Photoshop and After Effects and the generative AI workflows within Pencil Pro, significantly enhancing the efficiency of marketing campaigns. With this collaboration, Brandtech aims to streamline the creative process, allowing brands to create, deploy, and optimize content more effectively.

David Jones, CEO of Brandtech, said that 2025 will be transformative for marketing as generative AI becomes widely adopted. The partnership not only ensures brand safety and quality but also enhances productivity by allowing users to easily import and export Adobe files. With the integration of Firefly Services, brands can expect improved performance, with content creation being 10 times faster and at least 50% cheaper than traditional methods.

Manchester City is inviting its fans to participate in the design of its new third kit for the 2026-2027 season through an innovative AI-powered competition. In collaboration with Puma and the AI firm DeepObjects, the club has launched a text-to-image uniform generator that allows fans to generate their own designs. Participants can create an account, upload a prompt describing their vision, and select from styles like "abstract" or "emotive brushstrokes."

Once users generate their designs, they can customize various elements, including collar styles, trim colors, and badges. Each participant receives 15 credits to create designs, although only two creations can be submitted for the contest, which runs until December 20. The winning design, determined by fan ratings and evaluations from Manchester City and Puma experts, will be worn by the players and sold to fans.

This initiative not only showcases the integration of AI in sports merchandising but also emphasizes fan engagement. Notably, players such as Ederson, Stefan Ortega, and Rico Lewis have already used the Puma AI Creator to design a goalkeeper kit inspired by the net of a football goal, marking it as the first AI-generated soccer uniform to be worn in a match.

→ Try the experience here

_{Credits: Deep Objects / Puma / Manchester city}

Coca-Cola has launched an interactive campaign featuring virtual snow globes, using AI to enhance audience engagement during the holiday season. This initiative is part of Coca-Cola's "Create Real Magic" campaign and is being showcased in an immersive format at Piccadilly Lights in London, powered by Ocean Outdoor’s DeepScreen® technology.

The campaign invites audiences to create their own traditional Christmas snow globes by sharing holiday memories with Santa Claus through an AI interface. The interactive digital artwork features Coca-Cola’s iconic Sundblom Santa Claus and showcases intricate 3D animations, including swirling snowflakes and detailed reflections. Viewers can scan a QR code to engage in real-time conversations with Santa and generate personalized snow globes based on their favorite holiday memories.

_{Credits: Coca Cola / Accenture / Ocean outdoor}

Northern Data Group has premiered The Glass Hermit, an AI-enabled short film directed by Oscar-winning artist Shona Heath. This project represents a groundbreaking fusion of art and technology, showcasing how AI can enhance human creativity. The film premiered in high-profile locations, including Times Square in New York and Piccadilly Circus in London.

The Glass Hermit was produced in collaboration with HELO, a creative production studio, and showcases the stunning visual capabilities made possible by AI technologies. Shona Heath, known for her work on the Oscar-winning film Poor Things, used AI throughout the production process, creating her own AI environment to manipulate original moving imagery.

The project aligns with Northern Data’s "Innovation Bravery" campaign, which celebrates creative pioneers like Heath who challenge the boundaries of artistic expression. The film balances experimental filmmaking with the latest advancements in AI, emphasizing the irreplaceable role of human creativity. Northern Data COO Rosanne Kincaid-Smith highlighted the importance of collaboration between human artistry and AI capabilities, promoting sustainability in the creative process.

Through the use of innovative techniques and collaboration with Emmy-winning production teams, including HELO and motion design collective Blind Pig, The Glass Hermit exemplifies how AI can drive artistic innovation while fostering discussions about the future of creativity in the digital age.

_{Credits: Shona Heath / Helo / Northern Data Group / Blind Pig}

Let’s dive into how to use it and test how it compares to the competition👇

Let’s begin with some tricky prompt

We’ve chosen to test the “storyboard” feature at the same time. This feature lets user sequence the action into multiple scenes to get more control over your output.

Sora divided our initial prompt into two scenes, each one detailing the subject, action and movement.

The prompt is well respected, we got the santa claus doing a backflip with his skiis in the alps, but, the physics looks off, and the overall result is far from the promises from their state-to-the-art video. Maybe our initial prompt was a bit difficult. Let’s try the Image-to-Video feature with a photo of a dinosaur previously generated in Midjourney.

The conclusion is the same, the legs physics seems off and random, but the overall movement is not bad. Let’s do a last test before we get to our final conclusion.

We tried to prompt a zoo tycoon video game action gameplay. (with way more details than the first test)

It does look good, the game interface is not bad, the scene stays consistent even when the camera is rotating around. This is our best result so far.

Sora, while promising, still has room for improvement. The physics and overall quality of the generated videos, especially in complex scenarios, are not quite at the level of the hype. However, the potential is clear, and with further development, it could become a powerful tool for content creation.

👋 See you next week for a new tutorial

🎉 Get The Mahazine Issue 3! 🎉

What Awaits You Inside?

🔥Global Trends: Stay ahead with insights into the evolving AI creative landscape.
👀Visual Inspirations: Fuel your creativity with captivating imagery.
📣Trending AI News: Keep up with the latest digital campaigns and advancements.
🎒AI Tutorials: Master new techniques and enhance your skills.
⭐And More...: Explore a wealth of resources tailored for creative minds

We truly hope you will enjoy the insights and inspiration we’ve packed into this edition. Your support means the world to us!

Thank you for being a part of our journey. We can’t wait to hear what you think!

Best,
The Mahazine Team