ChAIos Media News - Back to Normal (sort of) edition

Hello and welcome to the 2nd ChAIos Media newsletter of 2024. The short lull between Christmas and New Year was clearly just a calm before the storm because it has been one heck of a week. As usual we will not cover every single detail of what happened this week but instead focus on what we believe matters the most to creatives who use AI.

Amazing creations from the community

As always, a lot of amazing stuff has been created this week. Here are a few picks.

The Lost Father by The Reel Robot
1970’s Los Angeles by KamaniMadeThis
Stunning reel by Nicolas Neubert
The Cult by our own Marco

New in generative AI this week

A lot has happened this week. Here are some of the major highlights!

The OpenAI GPT Store launches this week!

Big news this week. OpenAI has announced that they will launch the eagerly anticipated GPT store as early as this week. This is an incredibly exciting development because everyone can now create GPTs based on their unique expertise and monetize them. The GPT store would have been live already if it weren’t for the fact that OpenAI spent some time firing and re-hiring their CEO Sam Altman! This time (unexpected firings aside) it’s going to be launched. Marco has created a very useful GPT called ‘Persona Personified’ with which you can define a design / marketing persona and then talk to it as if it were a real person.

Check out Persona Personified (chatGPT Plus subscription required)

Midjourney updates their v6 model and splashes around in hot (copyright) water

Midjourney has updated their recently launched v6 alpha model a few days ago. Improvements include improved aesthetics, prompt coherence, and image quality. Needless to say, the quality of the v6 model is quite amazing.

At the same time there’s a growing amount of turmoil going around surrounding Midjourney and copyright issues. The issues are mostly brought up by folks who, let’s say are not as big a fan of generative AI as we all are but one thing needs to be said: there is some merit to some of the issues that were brought up. Since these issues are quite complicated as well as nuanced, we decided to dedicate an entire X space to it last weekend. It was an extremely interesting and constructive discussion. This space featured the amazing KamaniMadeThis as our co-host as well as a whole bunch of extremely interesting speakers. Check below to listen to the recording.

RunwayML announces the second edition of GEN:48

From February 3 till February 2 it’s on again. The RunwayML GEN:48 AI movie competition. As far as we are concerned this is still by far the most exciting AI competition there is. What’s not to like about blocking off an entire weekend to try and create an amazing piece of AI cinema? For those who are curious what to expect, head over to the finalists page from the first GEN:48. Since the technology has improved quite a bit since the last edition we are expecting to see some true gems coming out of this competition.

Pika Labs goes subscription based

We all knew this momeny was coming. Pika Labs has introduced subscription plans for their service. GPUs are pricey and therefore the era of free Pika Labs renders is coming to an end. Or well, almost.

It is still possible to generate a limited amount of free videos with Pika Labs at least for now. On the free plan you will be able to render 300 videos. If you need more you will need to sign up for a plan. The standard plan is very affordable at $10 per month and will give you 1050 credits as well as watermark-free videos, a feature a lot of Pika users have been waiting for. Not a bad deal if you ask us.

For the heavy users there’s a $60 per month plan offering 3000 credits per month. This may sound like a steep price difference but the $60 plan also continues unlimited ‘Chill’ generations which means there is no limit to how many videos you can render. They will simply take a little bit longer to finish.

Recent movers and shakers

Assistive Creative Partner Program

Assistive is one of the newest kids on the block in the AI video arena. They have recently launched a creator’s program for which you can apply. In case you’re admitted you get a free subscription, lots of credits and more.

50% off a SplashMusic subscription through AmliArt

The best deal we’ve seen yet. Use the link through AmliArt to get 50% off a plan. SplashMusic generates some excellent music you can use in all your AI projects.

Magnific.AI drops an amazing AI image upscaler update

It’s no secret that we are huge fans of Magnific.AI here at ChAIos Media. While Magnific.ai was impressive even in it’s existing form, Javi and his partner in crime have done it again. The new version of the upscaler is able to create images at a staggering 10,048x10,048 pixels. We are not exaggerating when we say that the image quality produced by the new pipeline is nothing but mind blowing. Small pictures don’t really do it justice. You’ll have to see some full scale examples to appreciate how amazing this is.

Congratulations Javi Lopez, you guys have done it again.

If you are serious about getting the absolute best possible quality for your AI art there’s nothing like Magnific.ai.

Apple announced that their Vision Pro release date is set for February 2nd in the US.

Apple Vision Pro and spatial computing are reshaping the AI creative landscape. By blending AI, computer vision, and extended reality, spatial computing delivers immersive experiences that merge the virtual with the physical world, innovating AI applications. This shift hints at moving from smartphones to wearable technology, enhancing real-time interaction and interpretation of the world. Considering grabbing one of these devices, or are you already exploring with AR/VR gear like the Quest? Do you plan on making content for the platform? Let us know on X. Here is Apple's Tim Cook's release date announcement:

Runway is testing user-created Custom Models for their Text-to-Speech tool.

Back in mid-December, Runway released their Text-to-Speech tool featuring dozens of voices. The quality is outstanding, and the tool is user-friendly, making it Ben's preferred TTS choice. Just last week, they introduced a beta version of their Custom Voice clone tool. Beta testers can upload a voice file or record directly in the web app to train a Custom Voice for use in the Text-to-Speech tool. Ben tried this with his own voice, creating a narration for his demo video. You can check it out here:

And that’s a wrap!

And that’s it for this week’s edition of ChAIos Media News. It’s been a crazy week and we’re pretty sure it set the tone for all of 2024. If you have a tip, you’d like to be featured, interviewed or you have anything else you think should be in next week’s edition, feel free to ping Marco or Ben on X or email us at [email protected].