Do You Use AI for Work? Here Are Five Legal Mistakes to Avoid

Introduction

Artificial intelligence (AI) innovation comes with a lot of legal complexity. Regardless of your industry, if you use AI for work, it’s critical to understand the legal implications. From copyright controversies to data privacy conundrums, the challenges can seem daunting. But they don’t have to be. This post will explore five key lessons for navigating the legal landmines of AI.

*DISCLAIMER*
While this post discusses legal topics and theories, it is not legal advice. Please consult a lawyer if you need legal counsel.

1. The Lessons of the Monkey Selfie: Who Owns AI-Generated Content?

One of the most curious legal cases, Naruto v. Slater, offers a surprising precedent. Naruto, a macaque monkey, famously snapped a selfie using a Slater’s (a photographer) camera. You can see the photo below. While the image garnered global attention, the court ruled that non-humans can’t hold copyrights. The photo was deemed public domain, leaving Slater unable to claim exclusive rights.

The implications for AI are profound. Just as a monkey cannot claim ownership, AI-generated content might not be copyrightable either. In Thaler v. Perlmutter, the U.S. Copyright Office rejected the copyright application for an AI-generated artwork, citing the lack of human authorship. This ruling suggests that any content solely created by AI may enter the public domain—a major risk for businesses relying on AI to produce logos or other critical IP assets.

Key Takeaway: Content created entirely by AI likely has no copyright protection. However, arranging AI-generated elements into a cohesive work, like a book or website, may grant some legal safeguards.

2. Data Privacy: Handle with Care

AI tools thrive on data, but mishandling sensitive information can lead to legal trouble. Data privacy laws vary globally, and non-disclosure agreements (NDAs) are the most universal safeguard against accidental leaks.

For instance, uploading confidential documents to AI platforms like ChatGPT might inadvertently breach NDAs. Many AI models store and train on the data they receive, meaning sensitive information could resurface in unexpected ways.

Pro Tip: When in doubt, don’t upload sensitive data to AI platforms without explicit permission from the data owner.

Key Takeaway: Protect yourself by keeping sensitive or confidential information off AI tools unless explicitly allowed.

3. Can AI Use Your Content Without Permission?

The legality of AI training on publicly available data has been fiercely debated. In Field v. Google, the court ruled that Google’s caching of websites didn’t violate copyright law because the website owner could have blocked the bots. Similarly, AI models likely have the legal right to train on publicly available content unless explicitly blocked.

But there’s a catch: when an AI-generated output closely resembles its training data, things get murky. In Anderson v. Stability AI, artists argued that Stable Diffusion’s outputs bore “substantial similarity” to their original artworks. While courts allowed copyright claims for the artwork, they dismissed claims over “style” likeness, highlighting how difficult it is to protect artistic influence.

Key Takeaway: While AI can likely train on public content, you might have recourse if the outputs are too similar to your work. Blocking AI crawlers from your site is possible but may harm your visibility in future AI-driven searches.

4. Beware of Publicity and Name-Image-Likeness Issues

AI-generated impersonations—whether visual, auditory, or textual—can land you in legal hot water. A notable example involves an AI-generated track by the artist “Ghostwriter” mimicking Drake’s voice. Despite its popularity, the song was quickly pulled due to violations of Drake’s right of publicity.

This extends to visuals as well. Tools like MidJourney or Stable Diffusion that generate images “in the style of” specific artists might cross legal lines, depending on the context and jurisdiction.

Key Takeaway: Avoid prompting AI to imitate a specific person’s style or likeness without their permission—it’s not worth the legal risk.

5. How You Can Shape the Legal Future of AI

As lawmakers struggle to catch up with AI’s rapid development, the landscape remains patchy. Over 150 AI-related bills are pending in Congress, with additional measures proliferating at the state level. However, this patchwork approach could lead to confusion and uneven enforcement.

You have a role to play. By contacting representatives and voicing concerns, you can help shape fair and consistent AI laws. The choices we make today will set the precedent for how future generations use and regulate AI.

Key Takeaway: Participate in policy discussions to help build a clearer, fairer legal framework for AI.

Final Thoughts

Artificial intelligence is a transformative tool, but its legal landscape is still evolving. Businesses and creators must tread carefully, understanding both the opportunities and risks. From safeguarding sensitive information to navigating copyright challenges, staying informed is your best defense.

As we embrace AI’s potential, we can’t forget that the responsibility to use it ethically and legally rests with us. You can learn more about AI’s legal risks int he video below!

About Daniel Russell

Daniel is a Partner at Go Fish Digital and a licensed attorney. During his marketing career, he’s helped structure market research programs for global companies like Whirlpool and Zynga, managed SEO audit projects for Wikipedia.org and The New York Times, and consulted General Electric and Monster Beverage Corp on their online ad campaigns and web presence. Daniel is frequently invited to speak on topics such as CRO, Reddit, and A.I. at the marketing industry’s top conferences including MozCon, Internet Summit, Google I/O Extended, and INBOUND. He’s also been a featured expert in interviews for Search Engine Journal and the Time for Marketing, Growth Hacking, and Marketing Agency Leadership podcasts. Before joining Go Fish Digital, Daniel was the founder of Attentiv (a communication software), a law clerk at Magruder (a corporate litigation law firm), a sales manager at Qualtrics (an XM software, now part of SAP), and a risk management analyst at Marubeni-Itochu (a Japanese trading conglomerate). Daniel was a Fulton Scholar Award recipient at Brigham Young University where he graduated with a degree in Economics and History. He later received his J.D. from the George Mason School of Law where he was the Editor-in-Chief of the Journal of Law, Economics & Policy. Daniel was admitted to the Maryland Bar in 2015.