Howdy Reader!

First of all, during this season of egg nog imbibing, Christmas cookie baking, last-minute present panicking, and fruitcake regifting, I want to pause and say thank you!

Thank you for being a part of this newsletter community. One of the best gifts is opening my computer the day after an email goes out and getting a chance to respond to people who have replied to the newsletter. Thank you for taking the time to chat. It means a lot.

Now, on to some cool new tools I had a chance to play with this week.

​Sora has launched!

Well, Sora-light has...

Come to find out, the model we’ve been seeing in the presentations is not quite the same one we got. The one that most people have access to is a stripped-down version that has some serious guardrails on it.

Nevertheless, there is a certain magic to it when it works right.

As you can see with this Santa video, it would be hard to believe that this wasn’t a stock video; in my opinion, it's pretty impressive.

Being somewhat of a beardologist, I’m always keeping my eyes out for facial hair that looks real in AI-generated content. And Santa’s beard in this one looks pretty good! Notice the unevenness of his mustache! To me, that hint of realism sells it.

I also like that the default settings provide two videos per prompt. You can change this to one, which will cost fewer credits, but two options can help you refine the prompt more straightforwardly.

Ok, Jeff, what don’t you like?

The major guardrails and general wonkyness...

One of the things I like about other video AI generators, like Kling or Runway, is the ability to upload your own images and animate them. You can do that with Sora, but right now, with a significant limitation.

Because they want to make it difficult for people to create deepfakes (I applaud them for this), you can’t upload images containing people to animate.

For example, I couldn’t upload my OWN photo of Santa I created using my Holiday Midjourney codes. But I could with this still scene of a snow-covered Christmas village.

​

I understand, but it's still pretty frustrating, especially since so many other AI platforms allow it.

Final verdict: Very cool, but it’s an arms race in the AI video space right now. If they want to keep up, they need to release features faster and safely relax some of the guardrails.

Second cool piece of AI news!

You probably know I’m a power user of Midjourney for creating images, and they just unveiled their biggest update in quite some time.

Here’s what’s new:

• Multiple profiles: You can now create a switch between multiple profiles.
• Mood boards: You can upload or add images to shape the model's look and feel in these new profiles.
• Custom Style: You can use those mood boards to create a tailored or broad style.

​

This means that Midjourney allows creators even more control over creating AI-generated images. We can also share our blended styles from our mood board with a code.

I’d love to know what you think of these updates and whether you will use any of them in your workflow. Hit reply and let me know!

It’s an exciting time to be a creator. I can’t wait to see what’s next!

See ya around the interwebs.

-Jeff