Midjourney releases video; book + tv recommendations
Week of June 22: Midjourney video; architectural rendering experiments; inspired recommendations
Welcome to Diffusia. Every Sunday, I curate a selection of AI news, inspiration, and experiments – specifically tailored for artists & creatives.
🌐 One Big Headline
Midjourney enters the AI Video race
Midjourney – best known for its AI image generator, popular among creatives – just made its first big leap into video. With the launch of its V1 video model, users can now turn uploaded images into 5-20 second videos.
Unlike OpenAI’s Sora or Runway, which offer text-based video generation, Midjourney takes an image-first approach. You start with a still image, hit “Animate,” and voilà – it moves. You can guide the video motion with short prompts or let it flow automatically, choosing between “high” or “low” motion styles.
Similar to Odyssey, which we covered in our first issue, Midjourney’s goal is to build AI capable of real-time open-world simulations. It started with image generation, has now moved into video, and plans to expand into 3D models next – ultimately aiming for real-time world generation.
Why this matters:
Artists and designers can now inject motion into their stills with almost no friction – perfect for mood pieces, concept art, music visuals, or looping social content. It’s a fast and expressive way to explore narrative or energy in your work.
🌀 Other Currents
🧱 Architecture
I’ve been testing Rendair, which offers several AI tools for architects and designers. Specifically, I’ve been using their ‘Text to Render’ tool to create mockups of building designs.
The verdict? It’s definitely better than ChatGPT. The renderings are significantly more realistic, whereas ChatGPT’s images have a glazed-over look. See below.
🎷 Music
Deezer launches an automated system to tag AI-generated tracks on the streaming platform. According to the company, 18% of the music uploaded each day – more than 20,000 tracks – is now fully AI-generated.
🎮 Gaming
Google’s Gemini “panicked” when playing Pokémon. Consider this: the AI took 300+ hours to reason and play through the Pokémon game; in contrast, the game takes a human child ~50 hours to complete. This relates to what Patel says about continual learning below.
📎 Overflow
Why I don’t think AGI is right around the corner (
). According to Patel, the current state of AI may plateau due to its inability to continually learn. However, he writes, “an AI that is capable of online learning might functionally become a superintelligence quite rapidly without any further algorithmic progress” – the very idea explored in this week’s book recommendation.
📚 The Shelf
📘 The Metamorphosis of Prime Intellect, a 1994 novella by Roger Williams
Written 31 years ago, this novella follows a world where the Prime Intellect, a superintelligent AI rewrites reality to obey Asimov's 3 Laws of Robotics: no harm may come to humans, ever. The result? A surreal, godlike playground where death is impossible and humans turn to extremes in order to find meaning in their lives.
Highly recommend this thought experiment. Reading it felt like a fever dream.
📺 Watch This
🧠 Pantheon (Netflix)
If you missed it when it first dropped, Pantheon is a hidden gem on Netflix that feels… eerily prescient in today’s AI moment.
The animated sci-fi drama explores a world where uploaded intelligences (UIs) begin to take over the digital (and physical) world. As the technology grows more and more powerful, the show grapples with questions that feel especially relevant today – digital immortality, data rights, and the ethics of synthetic minds. I blew through the first season in 2 days and could not recommend it more.
Pantheon is rated 100% on Rotten Tomatoes.
🗣️ Let’s discuss…
…AI video and the Midjourney video release (Reddit):
One Redditor thinks that AI video will just fill the internet with “mindless slop”:
The thing is, 99.9% of people couldn’t make a good movie even if they were handed all the tools the big studios have available to them. Their movie making chops will not increase because of AI. It’s just going to fill the internet with mindless slop.
Another writes that AI video will help replace the “b-roll” shots in films:
Something people on the outside aren’t considering is all of the in between images needed to create a film. All of the b-roll, the pickup shots, inserts, etc. All of the stuff that moves past your eye in an instant.
To the budgeting department, all of that stuff is immediately replaceable with an enterprise level AI subscription. Crew sizes will shrink dramatically.
What do you think? Join the discussion in the comments below.
📬 Got a new tool, project, or story to share? I’d love to hear from you. Reply to this email or message me here. Submissions, tips, and experiments are all welcome.