Generative AI Is Heading Down A Dangerous Legal Path

There’s no denying the potential of Stable Diffusion, an artificial intelligence text-to-image model.

In some cases, the technology, released in 2022, appears to be nothing short of magic — it’s truly that powerful.

Don’t Miss: The House-Printing Robot Shaking Up a $7.28 Trillion Industry

With a simple text description, Stable Diffusion — owned by Stability AI — can quickly generate a high-quality image. This is possible because the software was trained on hundreds of millions of example images pulled from the online world. 

Despite potential ethical and legal concerns, AI startups have gained significant traction over the past year. For example, ChatGPT parent, OpenAI, raised $10 billion from Microsoft Corp. as the tech giant looks to take on Google using AI integrations. RAD AI is currently raising over $3.5 million from over 3,300 retail investors for it’s AI marketing platform built to understand emotion. 

Where Do The Original Images Come From?

While many of the images that trained Stable Diffusion are in the public domain or published under permissive licenses such as Creative Commons, that doesn’t hold true across the board. Others are copyrighted, and as you can imagine that doesn’t sit well with the artists and photographers whose work is encroached upon.

In January, three artists filed a class-action copyright lawsuit against Stability AI. Shortly after, Getty Images Holdings Inc. — the world’s leading supplier of stock assets — followed suit.

Getty shared the following in its lawsuit:

“Stability AI has copied more than 12 million photographs from Getty Images’ collection, along with the associated captions and metadata, without permission from or compensation to Getty Images.”

On the surface, this type of copyright case appears cut and dried. Stable Diffusion used millions upon millions of images without permission or providing compensation. But with generative AI being such a new technology, the legal system is starting at ground zero when approaching these cases. 

To stay updated with top startup news & investments, sign up for Benzinga’s Startup Investing & Equity Crowdfunding Newsletter

What Are The Copyright Implications Of Generative AI?

There are several sides to this legal argument. On one side, some believe that copyright’s fair use doctrine should allow Stable Diffusion — and like companies — to use the images.

Conversely, there’s the argument that the company has violated copyright law on a large scale. And if that’s the case, AI companies are doing more than heading down a dangerous legal path. They must also consider the implication on their respective businesses, including the potential costs and the time it would take to obtain the required usage licenses. It’s also possible that all this red tape could make it nearly impossible for new entrants to the generative AI market to gain traction. 

Don’t Miss: Qnetic Unveils Revolutionary Flywheel Energy Storage System to Accelerate Renewable Energy Adoption

What’s Next For Stability AI And The Lawsuits?

Stability AI has yet to respond to the lawsuits. While no one knows the defense strategy the company will use, it may take a similar approach as Google when it ran into legal issues when scanning books in the mid-2000s. 

The search giant was sued by authors for copyright infringement, but the company was able to successfully argue that its scanning was fair use. A 2014 ruling in favor of Google said, “The result of a word search is different in purpose, character, expression, meaning and message from the page (and the book) from which it is drawn.”

Time will tell what happens with Stability AI’s legal battle. Either way, it’s sure to shape the future of the AI industry. 

See more on startup investing from Benzinga.

• Kevin O'Leary-Backed Startup Lets You Become a Venture Capitalist With $100

• The Startup Behind The Automated Future Of The Fast-Food Industry