"The AI Copyright Fight: A Guide to Common Misconceptions and Legal Gray Areas"

"The AI Copyright Fight: A Guide to Common Misconceptions and Legal Gray Areas"

In early 2023, the world witnessed a significant copyright battle involving AI image generators. Two parties, a group of artists known as Andersen et al. and Getty Images, separately sued AI image generators for copyright infringement. This event brought attention to the complex history and evolution of copyright laws and raised questions about the role of AI in creative industries.

To understand the roots of copyright, we must go back to the royal courts of Europe around 500 years ago. European monarchs began granting privileges and licenses to their favored artists, allowing them exclusive rights to publish and distribute their works. These privileges, known as "copyright," initially aimed to control the printing press, as only royally appointed publishers could afford the expensive equipment. These publishers adhered to censorship standards and faced consequences for printing heretical or seditious content.

However, the rise of Gutenberg's more affordable printing press in the mid-1500s disrupted the control of royal publishers. They struggled to keep up with the proliferation of unlicensed works, leading to a shift in public opinion. English poet John Milton advocated for the "liberty of unlicensed printing," which eventually influenced the passing of the Copyright Act of 1710 in England. This act granted copyrights to authors, marking the first law of its kind.

Over time, copyright laws evolved to adapt to changing times and technologies. The Copyright Act of 1976 introduced the concept of "fair use," allowing certain entities to use copyrighted materials for free under specific conditions. Initially, fair use was limited to nonprofits and government entities. However, courts began granting allowances for commercial use if other factors were compelling. For example, in the case between Playboy competitor Perfect 10 and Google, the court ruled that Google's search engine's use of thumbnails was "highly transformative" and fell under fair use.

The concept of "transformative" use has been central to many copyright lawsuits in recent decades. However, determining whether a use is transformative is subjective and decided on a case-by-case basis in court. Judges consider various factors, including how much the purpose or use changed in the reproduction and the content of the copying. For instance, if a work like "50 Shades of Grey" is so transformative of another work like "Twilight" that most readers don't realize the influence, it becomes challenging for the original creator to claim copyright infringement.

When creators adapt copyrighted material from one medium to another, such as film adaptations or fan fiction, the reproduction is often considered a "derivative work." If the adaptation does not change enough aspects of the original story, licensing fees must be paid. Fair use also protects parodies, criticism, commentaries, and reviews, as they often serve a public benefit and fall under the First Amendment.

The recent lawsuits around AI image generators revolve around two main issues. Firstly, these generators use copyrighted images without permission during their AI training. Secondly, they produce derivative versions of copyrighted work. While Getty Images claims that only some AI-generated images are derivative, Andersen et al. argue that all are. However, as we have seen, being derivative does not necessarily protect a reproduction under copyright law.

One of the plaintiffs, Andersen et al., takes their lawsuit claims even further by arguing that AI generators should pay to commission or license work from artists when they produce work "in the style" of a particular artist. If such precedents were set, it could risk infringing on copyright whenever creators take inspiration from others' work. This raises concerns about the potential for copyright to regress into a system that benefits a select few artists and enforces forms of censorship.

In light of these copyright battles involving AI, there are actionable steps that can be taken to navigate the legal and ethical gray areas. Firstly, AI companies should seek permissions from copyright holders before using their data to train AI models. While scraping publicly available data may not be illegal, using that data to train AI models raises ethical questions. Secondly, AI companies should give proper attribution to artists whose work has been used in the training process. Even the most relaxed creative commons permissions usually require attribution. Finally, AI-generated images that have not been modified by a person should remain in the public domain without copyright protection. Machines and algorithms are not humans and should not be granted the same creative protection rights.

In conclusion, the AI copyright fight sheds light on the complex and ever-evolving nature of copyright laws. The concepts of "transformative" use and "fair use" play significant roles in determining the legality of reproductions and adaptations. However, the subjective nature of these concepts and the lack of a clear line make it challenging for creators to know if their use qualifies as fair use. As AI continues to advance and play a larger role in creative industries, it is crucial for lawmakers and creators to navigate the legal and ethical implications to ensure a fair and balanced system that fosters creativity while respecting the rights of copyright holders.

Actionable Advice:

1. Seek permissions and licenses from copyright holders before using their data to train AI models.
2. Provide proper attribution to artists whose work has been used in AI-generated content.
3. Advocate for AI-generated images that have not been modified by a person to remain in the public domain without copyright protection.

Sources:

• "The AI Copyright Fight: A Guide"
• "Overrated vs. Underrated: Common Beliefs We Get Wrong"