Written evidence submitted by copyright alliance

gai0097

About the Copyright Alliance

The Copyright Alliance appreciates the opportunity to submit written evidence to the House of Commons Science and Technology Committee’s call for evidence on the governance of Artificial Intelligence (AI) (“Call for Evidence”). The Copyright Alliance is a non-profit, non-partisan public interest and educational organization representing the copyright interests of over 2 million individual creators and over 15,000 organizations in the United States, across the spectrum of copyright disciplines.[1] We are dedicated to advocating policies that promote and preserve the value of copyright, and to protecting the rights of creators and innovators who rely on copyright law to protect their creativity, efforts, and investments in the creation and distribution of new copyrighted works for the public to enjoy.

U.K. AND U.S. COMMITMENT TO STRONG COPYRIGHT LAWS

The dedication of the United States and the United Kingdom of championing and protecting creativity through copyright law is the reason why our countries are global creative and cultural leaders.[2] Strong copyright laws have been a successful driver of economies and innovation[3] and has fostered numerous, successful creative and cultural collaborations between the United States and United Kingdom. These have included telling stories that have captivated global audiences, such as the Harry Potter, Lord of the Rings, and the Star Wars series. Various music genres, such as rock-and-roll, jazz, pop, and musicians like The Beatles, The Rolling Stones, Simon & Garfunkel, Jimi Hendrix, and Adele have also benefitted from copyright protections in both countries. In addition, countless news content and scholarly journals facilitate the sharing of global scientific and general knowledge, which is possible because of the strong copyright laws in both our nations. AI and related technologies, which are evolving at a rapid pace, can be complementary to the benefits afforded by strong copyright laws. However, the U.S. and U.K. must be thoughtful in its approach to healthy AI innovation when examining the application of AI in a copyright law context.

The Copyright Alliance supports the responsible development of AI technologies as the creative communities are immersed in the development and use of AI. Thus, we appreciate that policymakers across the globe, including the U.K. Parliament’s Science and Technology Committee, are carefully examining the effects of AI on a wide range of stakeholders, including the creative community. During any analysis, it is vital for policymakers to preserve the integrity of intellectual property (IP) rights and markets. IP is an engine of economic growth and creativity, not a barrier, spurring healthy innovation and progress in the arts and sciences. Any formulations of new AI laws and policies must respect IP—and in particular copyright law. Copyright law is the foundation through which independent to large-scale creators, rightsholders, and innovators are empowered to craft and author inspiring, innovative, and pioneering works, and drives innovation, including in the AI space.

THE CREATIVE COMMUNITY DRIVES AI TECHNOLOGIES

The creative community fuels the development and use of AI in myriad ways. Creators and copyright holders have actively developed or utilized AI technologies as part of their creative endeavors to produce or enhance creative works. For example, AI technologies have been developed or used by video game developers,[4] journalists, software coders, visual artists,[5] and musicians.[6]

The creative community also drives AI innovation and development by creating and disseminating high quality copyrighted works that AI developers use to train AI machines. The reason that the creative community, from independent to large scale creators and rightsholders, is able to create such high-quality works is because of strong copyright laws. And it is these high-quality works that are ideal to train AI machines to generate high-quality output, including through techniques such as text-and-data mining (TDM).[7]

The use of the creative community’s copyrighted works for AI training purposes should be subject to marketplace licenses. Indeed, there are already organically developed and robust licensing markets for use of copyrighted works as training materials for AI, and it is essential that these markets are preserved and protected. For example, publishers frequently license TDM of journal articles and other publications.[8]

Promoting strong copyright protection for creative works and marketplace licensing for the ingestion of those works for AI training further drives AI innovation.  Strong copyright laws and marketplace licensing ecosystems incentivize creators and rightsholders to lawfully enhance and catalog copyrighted works—such as through semantic enrichment, metadata tagging, content normalization and data cleanup—which makes the copyrighted works even better suited for developing AI applications.

Keeping in mind the importance of protecting and preserving the rights of the creative community through strong copyright laws and the vital, complementary role it plays in AI development, we answer the following questions below from the Call for Evidence.

1.      How should decisions involving AI be reviewed and scrutinised in both public and private sectors? Are current options for challenging the use of AI adequate and, if not, how can they be improved?

Protecting copyright promotes AI by encouraging investment in the creation of high-quality works that can, in turn, be licensed in the marketplace to help train AI. As explained above, creators and copyright holders provide a variety of products and services that are vital to the development of AI technologies. Thus, when decisions involving AI are reviewed and scrutinised in both public and private sectors, copyright laws should not be cast aside or demeaned in favor of new policies— especially policies which obligate the creative community to subsidize AI technologies under the misguided belief that doing so is necessary to incentivize AI development.

In terms of current options for challenging the use of AI, there are best practices from corporations, research institutions, governments, and other organizations addressing transparency around AI training data. It is vital to encourage these transparency practices around AI training data which include recording what works are ingested by AI systems and for what purpose, which helps to ensure that copyright owners’ rights are respected. Infringement analyses, copyright defenses, and licensing terms disputes can all benefit from transparency best practices, and they can also be crucial in promoting safe, ethical, and unbiased AI systems. On a policymaking level, we urge the U.K. government to continue engaging all AI stakeholders to make evidence-based decisions and to encourage existing best practices to address AI use of copyrighted works.

2.      What lessons, if any, can the UK learn from other countries on AI governance?

For many years now, lawmakers and policymakers in a number of countries, including the United States, have been carefully examining the intersection of copyright law and AI and the implications of this rapidly evolving technology.[9] Even as a leader in AI technologies, the United States has not deemed it necessary to enact or recommend any new exceptions to copyright law for AI purposes. And for good reason: licensing to support AI application is robust and absent contrary evidence there is potential for significant harm by prematurely legislating in this rapidly changing arena. Additionally, arguments that unauthorized AI uses of copyrighted works (including TDM uses) are generally permitted under U.S. copyright law, particularly on fair use grounds, are mischaracterizations of U.S. law. There can be no broad characterizations of AI exceptions in the United States because our copyright laws do not provide for any such broad exceptions. To the extent that the United States is a model of innovation in this area, it is because its AI licensing markets have been allowed to organically develop.

The United States is not alone in its treatment of the AI licensing market and its relation to copyright law. Other countries that have considered AI regulations and policies with respect to copyright laws include Hong Kong,[10] South Korea,[11] Australia,[12] and Canada.[13] Significantly, each of these countries has declined to take action, postponed decision making as premature, or otherwise not taken action. The United Kingdom should follow this model of monitoring, and permit the licensing marketplace to flourish rather than impose any draconian exceptions for TDM.  It should not follow the approach taken in Japan[14] or Singapore,[15] which are overbroad, lack evidentiary support for such broad exceptions, and ultimately disincentivize the creation of high-quality copyrighted works that could be used for AI ingestion.

CONCLUSION

It is critical to AI innovation that the creative contributions and impact of the creative community to AI are acknowledged. As the U.K. government studies AI issues, we urge U.K. lawmakers and policymakers to continue championing the rights in the underlying works which fundamentally shape, develop, and train AI machines—specifically, the copyrights and creativities of creators and rightsholders.

We are encouraged by this Call for Evidence and hope that the United Kingdom will continue its longstanding leadership role along with the United States as a champion of the creative community and the copyright foundations on which both of our countries rely. We thank the U.K. Parliament’s Science and Technology Committee once more for the opportunity to submit evidence. Please let us know if we can provide any additional input or information or address any further questions.

(November 2022)