As artificial intelligence grows closer to emulating human communication, creativity, and expression, it already has demonstrated one very human tendency: to repurpose the work of others. Safeguards must be put in place to ensure that AI-generated content does not infringe upon people's intellectual property. This work includes designing algorithms that actively avoid producing arguably 'copied' content as well as programming AI with consideration for privacy and individual rights. Additionally, transparency in how AI algorithms operate and what data they use is crucial in instilling confidence in their output. With the proper safety nets and measures in place, AI can become an invaluable tool for producing high-quality, original content that contributes to the arts and public discourse.

AI-generated IP Infringement

In cases where these safeguards fail or are not implemented, the law must determine who bears responsibility and how damages should be awarded. Hence, it is important to be aware of the potential risks of AI-generated content. For instance, AI-generated content could infringe on someone's intellectual property (IP) rights by reproducing copyrighted material. If this occurs and the content is used, someone may be held liable, forced to remove the content, and potentially required to compensate the copyright or trademark holder.

The question then becomes, which party is responsible for the infringement?

Depending on the situation and the jurisdiction, liability may rest with either the creator of the AI program, the platform, the content publisher. The creator of the AI-generated content (that is, the person who told the AI what to create) might be liable if the algorithm includes plagiarized passages, uses a trademark without permission, or bases an artwork on an image or other media for which the output "creator" does not properly license from its owner.

In some cases, the creator of the AI content may not be the only party liable for any infringement or objectionable material presented within the AI's output. For example, if it can be shown that the art or text was created using copyrighted material used to train or inform the algorithm, partial liability could lie with the party responsible for creating the training database.

The US is one of several countries that have established copyright laws to protect publishers from copyright infringement by AI. South Korea and Japan have followed suit. Advocates of such regulation cite the need to encourage innovation in AI. By protecting publishers from liability, AI developers may be more likely to invest in the development of these technologies.

Proving the Case

To prove copyright infringement, the alleged aggrieved party must clear a number of hurdles:

  • Demonstrating that the defendant's work is substantially similar to the protected work. This is a legal term that means "more than mere coincidence." If two works are very similar, it's more likely that one has been copied from another.
  • Depending on the type of IP rights being claimed (such as copyright or trademark infringement), there may be additional evidence required in order for a successful case to be brought against an AI provider. This could include proof of access to the original work prior to its use by the AI or proof that the same creative process was used in creating both works.
  • Showing that the plaintiff suffered actual damages in the form of lost revenue as a result of the infringement.

Relevant Case Law

The legal landscape surrounding AI and copyright is still evolving, and the laws governing liability for AI content are still developing. To fully protect themselves, organizations using AI to create written content or art should consult with a suitably experienced attorney to minimize their exposure to liability. Similarly, a lawyer specializing in AI and digital technologies can assist anyone who believes their IP has been misappropriated - whether by AI or human activity - in securing their rights, protecting their creative output, and securing damages if necessary.

Cases of AI-originated IP infringement form a relatively new and evolving area of the law, and there have been only a few instances where courts have addressed such issues. However, some recent cases may provide some indication as to how courts may rule in future cases.

In one US case, a photographer sued a website that used AI to crop and display his copyrighted images without his permission. The court held that the website was liable for copyright infringement because it had not obtained a license to use the copyrighted images, and its use of AI to crop and display the images did not excuse its infringement.

In another case, In 2020, Shein, a Chinese fast-fashion giant, faces dozens of lawsuits alleging design theft. Stussy, an Irvine, Calif.-based streetwear company, filed a lawsuit in March 2021 alleging Shein sold shirts and shoes bearing nearly identical copies of its logo without permission. The lawsuit claims trademark infringement and counterfeiting, trademark dilution, and unfair competition. Jennifer Stark, a Los Angeles-based artist known for her intricately patterned murals, also sued the company, alleging it reproduced her work on its inexpensive skirts, swimsuits, and shirts.

To date, courts have generally held that the use of AI does not excuse copyright infringement and that organizations using AI must still obtain proper licenses and permissions to use copyrighted material. However, the issue of whether AI-generated content can infringe on someone's IP is still an open question, and it is likely that courts will continue to grapple with this issue as AI technology continues to evolve.

Legal Safeguards

An attorney with experience in AI-related legal matters can help organizations protect themselves from the consequences of using AI that generates incorrect, discriminatory, or copyright-infringing output in several ways:

  1. Drafting and implementing comprehensive terms of service agreements that clearly outline the intended use of AI-generated content and the responsibilities of the organization with respect to such content. This can help ensure that the organization is not held liable for any infringing or objectionable content generated by the AI.
  2. Navigating the complex and dynamic legal landscape surrounding AI and copyright and advising clients on best practices for obtaining proper permissions to use copyrighted material used in the development and training of AI algorithms.
  3. Implementing safeguards and ethical considerations to ensure that the AI-generated content is accurate and does not infringe on the rights of individuals or organizations.
  4. Establishing and registering valid copyrights.
  5. Setting up dispute resolution or taking legal action related to AI-generated content.
  6. Drafting user agand using AI and with appropriate precautions in place, AI has the potential to become an incredibly powerful tool for creating quality content that does not infringe on the intellectual property and rights of others.

    The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.