Generative AI and copyright: business use, risks, and contractual lessons

Generative AI has become a part of daily operations for many businesses. Texts, images, presentations, campaign materials, product descriptions, and internal documents are being created, in whole or in part, with the help of AI tools. At the same time, however, there are increasing misunderstandings regarding the copyright situation. It is a common misconception, for example, that content created with AI is automatically "free to use" or that its legal status can be treated the same as a classic work created by a human.

The situation is more complex than that. Current EU and Hungarian copyright logic remains human-centric, even as the business application of AI systems spreads rapidly. For businesses, the essential question is therefore usually not whether generative AI can be used, but how controlled, documented, and contractually regulated that use is.

"The legal risk of using generative AI often appears not in the output itself, but already at the input stage and throughout the entire chain of use."

What makes this a copyright issue?

When using generative AI, at least two separate questions must be distinguished.

The first question is whether the finished content itself can enjoy copyright protection. Copyright traditionally protects works that are the result of the creator's own intellectual effort. This logic is difficult to apply to cases where the substantive part of creating the content is performed by an AI system, and the human role is limited to prompting, selecting, or subsequent editing.

The second question is whether the use of a given output could infringe on the rights of others. In the case of corporate marketing materials, branding elements, or campaign visuals, the question is not only "who created it," but also how much the result resembles protected content, what sources it might be based on, and what terms of use apply to the platform in question.

In other words, from a business perspective, it is not enough to look at whether the tool "works well." You must also examine how legally defensible the use of the content produced in this way is.

The input side is just as important as the output

In corporate practice, the most sensitive point is often not the finished content, but what is fed into the AI system. Employees and external partners frequently upload materials as prompts or background information that are subject to copyright, trade secrets, know-how, or personal data.

This can pose a risk from several directions. It can be particularly sensitive if, for example, the following are uploaded into the system:

• third-party articles, images, or texts
• client materials, internal presentations, or draft contracts
• unpublished marketing concepts or creative briefs
• content for which the company does not hold full usage rights

In such cases, copyright, confidentiality, and data protection risks can easily become intertwined. Therefore, one of the most important practical questions regarding the use of generative AI is what materials a business allows to be uploaded into a given system at all.

"When it comes to the business application of AI, the real question is usually not whether it can be used, but how controlled and documented the company's use of it is."

Why is the AI Act important if the issue is fundamentally about copyright?

The AI Act is not a copyright reform, but it still has an important impact on the topic. EU regulation imposes specific obligations on providers of general-purpose AI models, including the development of policies to respect copyright and transparency requirements regarding training data. While this does not in itself resolve all tensions between generative AI and copyright, it clearly points in the direction that the legal background of models and the documentation of the usage chain are becoming increasingly valuable from a business perspective.

For companies, this is significant because the approach that choosing an AI tool is merely a technological or procurement decision will become increasingly unsustainable in the future. The platform's legal background, terms of use, level of transparency, and contractual guarantees are all becoming substantive considerations.

General IP clauses are no longer sufficient in contracts today

Generative AI is appearing more and more frequently in agency contracts, software services, development agreements, and content creation collaborations. However, many existing contracts still assume that the output is the result of a human creative process and do not explicitly address the specific risks of AI-assisted work.

It is particularly advisable to clearly address at least the following issues:

• whether generative AI can be used during the performance of the work
• whether the client must be informed about this
• who bears the risk of legal compliance and infringement
• what level of human review is mandatory before submission
• what warranties the service provider offers regarding the usability of the output
• what limitations or platform dependencies remain regarding the created content

In practice, these clauses are not just technical details. They are actually about how clear the situation regarding liability and risk allocation is in the event of a future legal dispute or business conflict.

What should you actually look at within the company right now?

Most businesses do not need a standalone "AI copyright project," but rather a review of a few well-targeted control points. Typically, it is worth reviewing:

• which AI tools have actually been integrated into daily operations
• what content is being created with these tools
• what materials employees are uploading into the systems
• whether there is an internal policy for AI usage and approval
• whether contracts address the specific risks of AI-assisted deliverables

In many cases, simply moving away from treating AI usage as an informal practice and instead documenting at a minimum which tools are used, for what purpose, and under what internal constraints can significantly reduce risk.

Closing thought

Using generative AI in business is not just a matter of technological convenience. Whenever marketing, branding, customer communication, creative content, or internal know-how are involved, copyright and contractual issues quickly come to the fore. The real risk is often not whether a company uses AI, but how well-considered, verifiable, and legally defensible the framework for that use is.

PLM Legal – Dr. Marcell Olajos Law Firm provides legal support to businesses in areas including commercial contracts, intellectual property, compliance, and technology agreements. If your company uses generative AI tools in marketing, content creation, internal operations, or the delivery of client projects, our firm can assist you in reviewing contractual risks, establishing internal AI usage frameworks, and resolving copyright issues related to business use.

‍