The nefarious use of digital replication has prompted the U.S. Copyright Office to issue a 72-page report detailing the need for more stringent protection of intellectual property, as AI-generated content continues to be used by scammers to deceive people in a myriad of ways, from convincing them to send off cash to help a relative, to influencing citizens through robocalls to reconsider their support of a particular candidate for public office.
On July 31, 2024, the Office published its “Copyright and Artificial Intelligence, Part 1, Digital Replications” report that addresses the need for additional federal legislation to protect individuals against unauthorized AI-generated digital replicas, including text, images, video and audio.
Digital Replications
Digital replications have become so fine-tuned, that even immediate family members are being duped into believing they are having phone conversations with their wife or husband, or brother or sister, or a close friend. And duplication of images and singing voices have become so exact that long-time fans find it impossible to distinguish between what is the genuine output of their favorite artist, and what is concocted through AI.
Now, the U.S. government must weigh how to best safeguard the public against such deception, while still allowing the technology to be used as a tool to invigorate creativity and innovation.
“As with all of the Copyright Office’s studies, our analysis is guided by the Constitutional goal of promoting creativity in order ultimately to benefit the public,” said Ms. Shira Perlmutter, the Register of Copyrights and Director of the U.S. Copyright Office. “This requires an appropriate balance, enabling technology to move forward while ensuring that human creativity continues to thrive. It is our hope that this Report will further productive discussions in Congress, the courts, and the executive branch, to help achieve that balance.”
In February, 2023, the Office reviewed a copyright registration that raised questions regarding the AI-generated content that was commingled with original, human-authored content. The work in question was a graphic novel that had images generated by the AI service Midjourney. The U.S. Copyright Office concluded that the novel was copyrightable work, but the individual images could not be protected.
“The topic of digital replicas does not fall neatly under any one area of existing law,” the Office stated. “While some characterize it as a form of intellectual property, protection against the use of unauthorized digital replicas raises overlapping issues including privacy, unfair competition, consumer protection and fraud. It relates to copyright in a number of ways: Creators such as artists and performers are particularly affected; copyrighted works are often used to produce digital replicas; and the replicas are often disseminated as part of larger copyrighted works. Moreover, the noncommercial harms that may be caused are similar to violations of moral rights protected in part through the copyright system.”
In March 2023, the Office released a policy statement, to essentially clarify its guidelines regarding AI-generated material. The office stated, “These are no longer hypothetical questions, as the Office is already receiving and examining applications for registration that claim copyright in AI-generated material. For example, in 2018 the Office received an application for a visual work that the applicant described as autonomously created by a computer algorithm running on a machine. The application was denied because, based on the applicant’s representations in the application, the examiner found that the work contained no human authorship. After a series of administrative appeals, the Office’s Review Board issued a final determination affirming that the work could not be registered because it was made “without any creative contribution from a human actor.”
In August, 2023, the Office published a Notice of Inquiry on AI and Copyright, seeking opinions on several issues related to AI-generated materials that replicate human work. The Office sought input on whether the U.S. Congress should enact a federal law that would protect against unauthorized digital replicas, including imitations of artistic style and voice imitations. By Nov. 15, 2023, over a thousand opinions flowed in, mostly from the public-at-large, with over 90% stating that new legislation is needed.
“The Office concludes that the time has come to adopt such a law at the federal level. Based on our analysis of the comments received, independent research, and a review of work being done at other agencies, we believe there is an urgent need for a robust nationwide remedy beyond those that already exist. In the sections below, we review the protections available under current laws and the gaps in their capacity to respond to today’s threats, explain the reasons for new federal protection, and provide recommendations regarding its contours.”
According to the U.S. Federal Register, Volume 88, No. 167, published on Aug. 30, 2023, the Office has a long history of scrutinizing issues involving machine learning and copyright. “The Copyright Office has long been engaged in questions involving machine learning and copyright. In 1965, the Office’s annual report noted that developments in computer technology had begun to raise ‘difficult questions of authorship’ – namely the question of the authorship of works ‘written’ by computers.”
The U.S. Congress is currently considering several bills related to AI-generated material, among them
H.R. 6943, the “No AI FRAUD Act.” The bill specifically relates to deepfake software, and refers to several examples. “On or around April 4, 2023, AI technology was used to create the song titled “Heart on My Sleeve,” emulating the voices of recording artists Drake and the Weeknd. It reportedly received more than 11 million views.” And “On or around Oct. 1, 2023, AI technology was used to create a false endorsement featuring Tom Hanks’ face in an advertisement for a dental plan.”
The bill, which was introduced on Jan. 10, 2024, refers to a Pew Research Center survey, that found that approximately 63% of Americans say made-up or altered videos “create confusion about the basic facts of current issues.”
The bill, if enacted into law, would impose several penalties for unauthorized distribution, transmission or other making available of a personalized cloning service, a digital voice replica or a digital depiction.
