Sponsored Content

The Intelligence may be Artificial, but how Real are the Risks?

Before diving into adopting AI–either by using an existing offering or creating a proprietary model, understand the related legal considerations.

  • By Elizabeth M. Stamoulis | Williams Parker
  • | 5:39 p.m. September 5, 2024
  • Williams Parker
  • Share

Making a logo. Summarizing a report. Businesses are starting to use generative artificial intelligence (AI) in many ways. But before diving into adopting AI–either by using an existing offering or creating a proprietary model–they should understand the related legal considerations, including:

1. Infringement–Several ongoing cases allege that (1) output created by AI infringes a third party’s rights and (2) the use of a third party’s materials to train AI infringes the rights of that third party. While we await the results of these cases, those using AI should be aware of the infringement risk of using AI-generated materials, and those creating AI systems should be aware of the risk of using third-party materials to train the system.

2. Ownership–In multiple cases, the Copyright Office has held that the input and tweaking of AI prompts is not “human” authorship that would allow for copyright protection in the resulting materials and has rejected copyright applications for this reason. Businesses creating important materials should be careful about using AI in the process.

3. Security and Use–In addition to any applicable laws, businesses should be aware that the terms of use for each specific AI offering can also govern the ownership and use of the content submitted to and created by AI. They should make sure that they are not entering any proprietary or confidential information into an AI system they do not control and that they are complying with the terms of use for the resulting content.

4. Privacy–Businesses should similarly be careful about entering any sensitive or personal information into a third-party AI system. They should make sure that they are familiar with the AI’s privacy policy and practices and that they are not violating any of their own obligations by inputting any information into an AI system.

5. Accuracy–Businesses should be aware that AI can sometimes “hallucinate” (make incorrect conclusions or assertions) and that no AI product is 100% accurate. Its output should always be carefully reviewed.

6. Disclosure–Some online platforms now require that any AI-generated materials posted to their site be labeled as such. Businesses should make sure they are complying with these developing requirements.

As you can see, while AI provides many opportunities, it also raises many questions. Businesses should take care to understand whether and how AI is being used in their business and how these risks are being addressed and mitigated.

Elizabeth Stamoulis is an attorney focusing on intellectual property practices. If you have any questions or need assistance, please feel free to contact Elizabeth at (941) 552-5546 or [email protected]