Embracing Authenticity over Automation

Decoding the WGA-AMPTP Agreement and the Unchanging Core of Creative Writing

The recent tentative agreement between the WGA and AMPT is cause for a celebration. Minimum writers rooms and pay raises for all. Even the AI bits are good:

If writers hired by studios want, they'll be free to "use AI when performing writing services, if the company consents and provided that the writer follows applicable company policies." But studios cannot require hired writers to use AI tools, and they must also disclose to their writers "if any materials given to the writer have been generated by AI or incorporate AI-generated material."

The cautionary reactions to these big wins brings to mind a recent article by an AI enthusiast who rejoices at the thought of a model replacing artists in the animation process:

Animation lovers, rejoice! The labor of love that is hand-drawn 2D animation is about to be revolutionized by a pioneering deep learning technique that automates the painstaking inbetweening process.

As someone who runs an AI-infused company, I understand the excitement surrounding the technology--BUT, as a former animator who worked several years as a professional inbetweener I can tell you the process is anything but "painstaking". Putting on headphones and drawing for 12 hours a day? To those of us who actually did it for a living--that sounds like paradise.

The current trend in the AI-space seems to be an abundance of swooning over the tech with absolutely zero idea of the passion and love that brought these creative endeavors into being in the first place.

Which brings me to, in particular the AI part:

Unsurprisingly, the AMPTP refused to give up its ability to train learning models on the scripts studios already own. But the WGA has reserved the right to "assert that exploitation of writers' material to train AI is prohibited by MBA or other law," suggesting that the organization intends to act as a litigious watchdog in the future.

Anyone with even a modicum of talent knows that it's not the words that make the writer but rather, the thinking behind the words. Training a model to write like you isn't training a model to write like you, it's training to write like your words.

Apparently, this needs to be repeated ad nauseam.

Subtxt (the application I built) stands ous from everything else in the market in that the application is trained on the underlying meaning of stories--not the actual words. No "plagiarism" here. Which is why all these "wins" are essentially pointless when it comes to writing great stories.

As with everything else, AI will only end the careers of those writers who believe that AI will end their careers.