AI Wants To Pick Their Brains, FTC Skeptical

In a memo to the U.S. Copyright Office, the Federal Trade Commission indicated that it was focusing on AI and intellectual property questions that go beyond the rights and liabilities under copyright law that are presently being litigated.

“The manner in which companies are developing and releasing generative AI tools and other AI products … raises concerns about potential harm to consumers, workers, and small businesses,” according to the memo.

The Register reports that the FTC is particularly concerned about assertions by corporations that using people’s content to train neural networks (a method that teaches computers to process data in a way that is inspired by the human brain) falls under fair use, which permits the unlicensed use of copyright-protected works in certain circumstances.

The AI companies claim they are making something novel and unique by transforming existing images and text. But according to the FTC, neural network-trained AI in particular “uses artists’ faces, voices, and performances without permission to make digital impersonations.”

It states that AI-generated content can mimic the style of specific creators to “exploit the name and reputation of the creator to gain sales and potentially compete with the creator.”

The memo also says that creatives complain that even when mechanisms have been implemented to offer consent and control over whether their work is used in AI training, the measures are “insufficient and ineffective,” nor is it always possible to tell whether their work has been included in AI training absent mandated disclosure.

The memo voices concerns about the ease with which AI-generated content can flood markets, “making it difficult for customers and other stakeholders to discern whether the content is AI-generated.”

The Register says the FTC is not focused directly on copyright enforcement, though copyright can be an element of its investigations.