In a recent blog, we opened the discussion of the rise of regenerative AI by entering the debate about the potential risks that are evolving as we speak. One of the risks, which is less mysterious and scary and is instead based on practicality, is the legal status of AI-generated products. Therefore, the following question is prevalent as AI generates more content in all phases and aspects of business.  Is AI product considered intellectual property, and can it be owned, copyrighted, and protected?

AI experts near me

The fact is that there is no established legality surrounding these topics at this time, and existing intellectual property (IP) laws may not even apply. This uncertainty translates into problems for those organizations seeking to protect their interests in AI-generated content. Therefore, businesses need to take proactive measures to mitigate potential risks, such as:

  • Ownership of AI-generated product: It is not clear who owns the works generated by AI, whether it is the provider of the AI platform, the individual who enters the prompts, or the creators of the original training data. In fact, no one is sure whether AI-generated work product can even be considered intellectual property – meaning it is uncertain if it can be owned or protected. In 2023, the United States Copyright Office offered guidance to determine copyright eligibility of AI-generated artistic works, stating that protection would depend on whether the works were the result of mechanical reproduction (response to text prompts) or if they were the creation of the author’s “own original mental conception”. The office also stated that well-known AI systems – such as ChatGPT, Midjourney, and DALL-E 2 – do not deliver work that can be copyrighted. This is due to the fact that users cannot exercise comprehensive creative control over how the AI generates material in response to prompts. That being said, modifications and innovative arrangements of AI-generated work may be copyrighted. The guidelines also stated that future copyright applicants must disclose if their work includes AI-created material. In addition, any previously filed applications must be amended to reflect that fact.
  • Illegal or unlicensed content in training data: Generative AI platforms sometimes include unlicensed or unlawful content in their training data, such as personal information, hate speech, or plagiarized material, which will likely inspire ethical and legal issues. Several jurisdictions have attempted to design a collaborative approach to deal with these types of concerns. Earlier in 2023, the UK Intellectual Property Office revealed their efforts to develop a code of practice for generative AI companies to mitigate their access to copyrighted data, followed by legislation if AI firms and creative sectors cannot come to a mutual agreement. The office further stated that the code of practice will offer guidance for AI companies to access copyrighted work as input, while safeguarding protections on generated output to protect right holders of copyrighted work.
  • Patentability of AI-generated created work: The computer scientist/inventor of DABUS (Device for the Autonomous Bootstrapping of Unified Sentience), an AI platform, filed patent applications in several countries for two inventions that he stated were the brainchild of DABUS – no human intervention was involved. In the patent applications, Dr. Stephen Thaler designated DABUS as the inventor and himself as the assignee of the rights. The patent offices of the UK, US, and EU all rejected the patent applications because “an inventor must be a natural person, not a machine” – and courts have subsequently upheld these rulings. They argued that a machine being named an inventor would significantly undermine the legal ownership and enforcement of patents. Dr Thaler has petitioned the US Supreme Court to review his case, and this issue is ongoing. 

Undoubtedly, the issues – whether legal, ethical, or practical – surrounding generative AI are still evolving and inspiring discussion. This means that companies that choose to rely on AI platforms for any substantive product at this time may yet face repercussions and unexpected problems. 

Alliance IT is a Sarasota-based managed services firm specializing in cloud services, security and compliance, and IT consulting across multiple industries.