So it begins.

This is something that I talked about a year and a half ago. AI was going to completely disrupt content creation. This means those who depended upon these jobs for their incomes were going to be in trouble.

I received pushback from those who do not believe that AI can replace human creativity. That was not easy to see at the time but should be fairly clear now. animation is at the tOP of the list because it is obviously the easiest.

It is also where the companies have data.

Hollywood Animation Being Crushed By AI

When we think of training models, the names of Big Tech come to mind. We have to consider, however, that models are emerging at every layer of the business world.

When we discuss the idea of AI models regarding studios, these are ones they train themselves. They utilize the data they have and feed it in. The result is a model that can be used for content creation.

Going back to the animation sector, these studios have, and are receiving, loads of animation content. This is owned by the studios, meaning they are able to utilize it as seen fit. The result is software that is increasingly able to handle animation tasks.

This will come in waves. Each generation of the models means the output improves. At the same time, the capabilities are expanded.

At this time, a few independent studios are working with it.

A few independent AI studios are positioning to produce animated content assisted by generative AI. Asteria Film is focused on making films, while Invisible Universe and Toonstar are launching original animated character IP on social media.

Asteria and Invisible Universe described their AI production workflow customizing a pretrained image or video generation model using a curated batch of original art assets to create a unique, custom model based on a specific character, franchise world or art style.

We can view these are the test cases. The major players are watching (likely doing something similar behind the scenes) as this is going to impact everyone.

These studios are able to generate characters where the IP is owned by them.

Sources described this process being done and seen as creatively viable for animation. In-house artists or animators develop a “core set” of original concept art representative of the original character or project. These assets form the dataset used to train any foundation image or video model the studio prefers (e.g., Stable Diffusion). The resulting fine-tuned model can then be used to drive subsequent content creation, whether producing outputs that replicate the studio’s specific characters or an aesthetic style present in the art assets.

These studio teams see fine-tuning as a way of executing on original IP developed in-house. Sources reflected that training custom models speeded and scaled artistic output while remaining visually consistent with the original IP or project.

Source

It is really that simple.

Larger Models Customized

The process is really one where an existing model is utilized, something like a Llama. This is the LLM which has the overall training on whatever the company behind it (in this case Meta) coudl get their hands on. As evidenced by the lawsuits that many companies such as OpenAi are facing, a lot of the training material was simply scraped.

Using the LLM as the foundation, in house data is added. This is called fine-tuning where new paramters are provided to the model.

From this point, new outputs can be generated. Since they are in-house, the LLM has no access (or rights) to them. This is a separate IP generated by the company that created the fine-tuned model.

job Losses

This is the future.

At this time, animators are not completely elminated. This is where the technical progression comes in.

We are dealing with technological capabilities of today. In 6 months, the tasks that it can hendle will be increased over today. That means what the computer can do will only keep growing.

In the meantime, another shift is taking place.

We are seeing animation production moving out of the Hollywood area. With improving tools, the capabilities of those elsewhere are closing the gap. This means that a studio can start to look at ways to offset the cost of production.

After all, if $25M or $50M can be wiped off the cost of a making a film, that is something the studio will welcome.

The other factor at play is the linked article. It was from Variety. That is not a technical publicaion nor is it industry specific. It is a mainstram site that covers many different topics. Yet the discussion of AI and animation is appearing there.

In other words, this is not a fringe discussion. We are looking at a massive shift in this technology. Many still feel it is a hype cycle yet the progress is telling us otherwise. We have evidence of companies, by name, in the movie industry already using the technology.

This is not something that is a forecast. It is actually happening.

---

What Is Hive

Posted Using InLeo Alpha