Blog | 11/10/2023

The Deepfake Conundrum: Balancing Innovation, Privacy, and Intellectual Property in the Digital Age

Team Contact: Benjamin C. Stasa

  • Artificial Intelligence
Share

The recent Hollywood strikes have brought to light significant challenges at the intersection of artificial intelligence, intellectual property rights, and personal agency. The use of AI in films to de-age actors or recreate past performances is becoming increasingly common, leading to concerns about control over an individual’s digital likeness.

This delicate issue bridges the gap between innovation, personal identity, and legal rights, and has heightened a broader conversation about the ethical implications of AI’s role in manipulating human identity. The strikes have uncovered the cracks in existing agreements and prompted a reevaluation of legal paradigms that have far-reaching implications, including the alarming rise of so-called deepfakes.

The Rise of Deepfakes

The use of AI in the entertainment industry is a complex issue that goes beyond mere contractual matters. As technology advances, the ethical considerations and challenges to traditional legal principles become more pronounced. The digital recreation of actors’ likenesses is just one part of this broader phenomenon.

Deepfakes, a technology that uses AI to create realistic-looking images or videos of people saying or doing things they never did, represent an escalation of the very concerns raised by Hollywood’s recent labor disputes. After months of negotiations, the strikes finally ended, with both actors and writers winning concessions and safeguards against the unfettered use of AI by studios, including, for example, the use of artificial intelligence in recreating actors’ performances.

The ability to manipulate digital likenesses has expanded beyond the movie studio, enabling unauthorized use of a person’s likeness with remarkable accuracy. The challenges faced by actors and writers in maintaining control over their digital selves is emblematic of a broader societal concern. Deepfakes threaten rights, such as copyrights and personality rights, by moving these issues from the big screen to our everyday lives.

Deepfakes and the Challenge to Copyrights

Deepfakes present a unique challenge to copyright law by enabling the unauthorized reproduction and alteration of copyrighted material. The use of deepfake technology to create manipulated versions of existing works implicates copyright owners’ exclusive rights, particularly the rights to reproduce, distribute, and publicly display their works.

Imagine a filmmaker’s work being edited using AI-generated content, transforming the narrative and characters without permission, or a musician’s voice and melody being used to create new songs. These examples illustrate how deepfakes can blur the lines of traditional copyright infringement, leading to complex legal disputes. In fact, these are not mere speculative concerns. In recent months we’ve seen songs that sound like the work or real artists generated by AI take the internet by storm. In one of the most high profile examples, the AI-generated song ​​”Heart on My Sleeve,” which simulated music by Drake and The Weeknd, went viral. In that case, Universal Music Group successfully invoked copyright violation to get various internet platforms to take it down.

Deepfakes raise questions about the nature of creativity and originality. If a deepfake alters a copyrighted work to the point where it becomes a new and original creation, does it still infringe on the original work’s copyright? The law is unclear on these matters, reflecting the need for legal evolution to address this novel challenge. As deepfakes continue to proliferate, we’re likely to see how courts address these questions in the context of current copyright law.

The advent of deepfakes has also sparked debate about potential fair use defenses, which are evaluated through a four-factor test under the Copyright Act. Could a deepfake that parodies or comments on a copyrighted work be protected under the fair use doctrine? Again, the legal landscape is uncertain, demonstrating the complexity of applying traditional copyright principles to this emerging technology.

Deepfakes and the Challenge to Personality Rights

Personality rights, protecting an individual’s likeness, voice, and other personal attributes, are also implicated by deepfake technology. Deepfakes pose a threat to these rights, enabling the unauthorized use of a person’s likeness with extraordinary accuracy. The ability to recreate and manipulate images of public figures and ordinary individuals alike for entertainment, promotion of goods and services, or disinformation creates another legal gray area that challenges traditional boundaries.

Consider a scenario where a deepfake creator superimposes a celebrity’s face onto a fictional character in a controversial or compromising scenario. The result is not only a tarnished image for the individual but also a complex legal problem that calls into question the efficacy of existing legal protections.

Existing personality rights laws often focus on unauthorized commercial exploitation of an individual’s likeness. However, deepfakes can be used for various non-commercial purposes, such as political commentary or satire. Moreover, deepfakes can transcend jurisdictional boundaries, adding another layer of complexity to legal enforcement.

Deepfakes and the Potential for Defamation

Beyond infringing on personality rights, deepfakes also present a substantial risk for defamation. By altering visual and auditory content, these AI-generated fakes can create entirely false narratives or misleading representations, leading to significant reputational damage.

For example, a manipulated video of a CEO making fraudulent claims can lead to not only public outrage but substantial legal challenges. Current defamation laws struggle to keep pace with this evolving threat, revealing a critical gap that requires legal and technological solutions.

The Current Legal Landscape

Deepfakes challenge legal systems across the globe. The current laws pertaining to this new technology are in their infancy, grappling with the complex intersection of innovation, personal rights, and IP law.

Some jurisdictions in the United States, such as Virginia, Texas, and California, have begun crafting specific legislation to address these challenges. Others rely on existing IP laws, common law and other statutory laws, attempting to apply traditional legal principles to an entirely new form of violation.

At the federal level, the issue of deepfake technology, while not specifically addressed, could be impacted by steps taken by federal agencies in response to President Biden’s recent Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. The Executive Order urges further study and consideration of “watermarking” AI-generate content, which could make it easier to detect deepfake content. Last summer, OpenAI, Microsoft, Google, Meta, Amazon, Anthropic, and Inflection made commitments to develop technology to clearly watermark AI-generated content.

Paths Forward

Adapting IP and other laws to address the unique challenges posed by deepfakes requires both creativity and caution. Potential strategies include:

  • Expanding Existing Protections: This could involve broadening the definitions within current IP laws to encompass deepfake-generated material, recognizing the unique nature of these infringements.
  • Developing New Doctrines: Creating specific legal doctrines that focus on the unauthorized manipulation of personal likeness through AI might offer targeted and effective legal remedies.
  • Encouraging Responsible Innovation: Collaborating with tech companies to foster responsible innovation, promoting transparency, and implementing deepfake detection tools could be a vital part of a comprehensive solution.
  • Public Awareness and Education: Educating the public on the risks and realities of deepfakes could form a critical aspect of a multi-pronged approach, enabling individuals to critically evaluate digital media.
  • Creating Effective Reporting Mechanisms: Establishing platforms for reporting and investigating deepfake incidents will enable a more proactive response, holding those responsible accountable.

Conclusion

The Hollywood writer and actor strike brought attention to the immediate concerns of AI in entertainment. However, the underlying issue of digital likeness control transcends the entertainment industry.

As technology evolves, so must the legal frameworks that govern its use. The task ahead is challenging but essential. By recognizing the multifaceted nature of this problem and pursuing comprehensive legal, technological, and ethical strategies, we can hopefully establish a system that protects individual rights without stifling technological innovation.

Keep Reading