Sora, Generate a Video of Someone Who Is Afraid of Getting Sued

Image via Twitter

“Publicly available videos.”

OpenAI CTO Mira Murati uttered those important words in wide-ranging interview with the WSJ’s Joanna Stern. They were discussing how Sora, the new text-to-video service, got its training data. When Stern pressed further, asking whether Sora used YouTube videos or Instagram reels, Murati made a face, above, like she just sat on something wet in the subway.

“Publicly available videos.”

As we approach the coming reckoning with the source of most LLM training data, the giants in GenAI are afraid. While they know that we, a bunch of yahoos making ‘Grams and YouTube videos won’t sue them, they’re definitely feeling the coming pressure from bigger players – including Alphabet, which owns Google and YouTube.

Learn AI Before It Takes Your Job

For those with busy schedules we offer a one-hour Beginning AI for Marketers, PR, and Journalists (March 21): a crash course meant to rapidly bring novices up to speed on using generative tools. Even if you think you know the basics of ChatGPT and other chatbots already, this class will improve your use of it with a focus on advanced prompting techniques, underutilized features, and a set of go-to tools for speeding up work.

Register

OpenAI is in trouble, and rightly so. Their current high-profile lawsuit with the New York Times is a rich vein of irony, showing how the company considers log analysis “hacking” and what Grey Lady really thinks of Open AI.

“The New York Times and its business model are built on world-class journalism,” write the NYT’s lawyers. “OpenAI and its business model are built on mass copyright infringement.”

Why is all of this important? Because while Generative AI (GenAI) is a valuable and significant tool, it is also built upon the work of millions of creatives. In the fast-paced, “move fast and steal things” ethos of Silicon Valley, this approach is ostensibly acceptable. However, we stand at the brink of continued, world-changing innovation that could unfold in one of two ways. In one scenario, Large Language Models (LLMs) collaborate with humans as part of a Centaur model, where the technology empowers humans, and they work together in perpetuity, side by side.

On the other hand, LLMs and AIs could replace humans entirely, leading to either a dystopian future filled with hungry, unemployed people or, if we adopt an optimistic viewpoint akin to Sam Bankman-Fried’s, a utopia characterized by leisure and slow-cooked soups. If I were a betting man, I’d wager on the vision of dystopia.

Murati, in this case, knows her company about to be pilloried. As of this writing, the benefits of AI tools – given all of their foibles, hallucinations, and failures – are new and interesting enough that we will give them a pass. But when those benefits are overshadowed by mass layoffs in media and marketing, people are going to start waking up. The robots took our jobs by passing off our work as their own.

That’s not right, fair, or sane.

And thus you get a CTO who is definitely afraid of what happens next.

We need your input for the Centaur Manifesto

Pete and I are working on a set of rules, a manifesto as it were, about AI and writing. We want to publish it as a something akin to the original Cluetrain Manifesto. It will contain a series of sections about how to ethically use AI, AI-generated art, and what AI companies need to remember when scouring the Internet for data. In short, we creatives need to plant our flag and take back control of this runaway technology.

We’d love your help. If you’d like to add a thought or two, please head over to this open Google Doc. It can me an immutable law, a rant, or even how you use AI in your own writing. Imagine you’re going to use this to teach future journalists and writers how to do their jobs in this changing environment.

Course Alert

For those with busy schedules we offer a one-hour Beginning AI for Marketers, PR, and Journalists (March 21): a crash course meant to rapidly bring novices up to speed on using generative tools. Even if you think you know the basics of ChatGPT and other chatbots already, this class will improve your use of it with a focus on advanced prompting techniques, underutilized features, and a set of go-to tools for speeding up work.

Register

Ready to start using AI like a pro?


Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.