Companies are increasingly using AI tools to create texts, visuals, software code, presentations and other business materials. However, AI-generated content should not automatically be regarded as free for commercial use or eligible for copyright protection. The use of AI tools raises important issues concerning authorship, rights of use, confidentiality, infringement of third-party rights and contractual liability.
Artificial intelligence has, in a relatively short period of time, become part of everyday business operations. Companies use AI tools to create texts, visuals, marketing campaigns, presentations, software code, analyses, internal documents and other business materials that are subsequently used in relation to clients, customers, investors or the public.
However, precisely because content generated by AI tools is increasingly used as a business asset, the question of its lawful use is becoming increasingly important. Companies should not assume that every piece of content generated with the assistance of AI tools is automatically “theirs”, free for commercial use and eligible for copyright protection.
The use of AI tools raises a number of intellectual property issues, particularly in relation to copyright, rights of use, infringement of third-party rights, confidentiality, trade secrets and contractual liability.
For companies using AI tools, the key question is no longer only what AI can produce, but whether such result can be lawfully used, protected, transferred to a client and commercially exploited without legal consequences.
Can AI-Generated Content Qualify as a Copyright Work?
Under Serbian law, a copyright work is an original intellectual creation of an author expressed in a certain form, regardless of its artistic, scientific or other value, purpose, size, content or manner of expression. The Intellectual Property Office of the Republic of Serbia also emphasises the element of an original intellectual creation of the author as the fundamental criterion for copyright protection.
This is particularly important in the context of AI-generated content, as it raises the essential question: who is the author?
The starting point is that, in legal terms, only a natural person may be an author. An AI system itself cannot be an author and cannot hold moral rights. Therefore, in the case of content created with the assistance of AI tools, the analysis should not focus only on the final result, but also on the manner in which that result was created, namely whether there is a recognisable, original and creative human contribution.
If a person uses an AI tool as a technical instrument, but independently designs the concept, structure and creative direction, and selects, edits, modifies and finally shapes the result, there is a stronger legal position to argue that, in the particular case, there is a human authorial contribution.
On the other hand, if a user enters a simple prompt and the AI tool independently generates text, an image, code or other content without any significant human creative contribution, the issue of copyright protection of such result produced by the AI tool remains legally sensitive.
In other words, not every piece of content created with the assistance of AI tools automatically constitutes a copyright work of the user. For copyright protection, it is not decisive merely that someone initiated the generation of content, but whether a sufficient level of human creativity and originality is reflected in the final result.
Who Owns Content Generated with AI Tools / AI Output?
In business practice, this question is often more important than a theoretical discussion of authorship. Companies want to know whether they may use an AI-generated logo, text, visual, software code or marketing material as their own asset.
The answer depends on several factors:
- the terms of use of the specific AI tool;
- the type of account or licence held by the user;
- the manner in which the output was created;
- the scope of human contribution;
- the contractual relationships between the company, employee, freelancer, agency and client;
- the purpose for which the content is used.
It is important to distinguish the right to use AI output from copyright in that output. The terms of use of a particular AI tool may allow the user to commercially use generated content, but this does not necessarily mean that such content is, in every case, the user’s copyright work, nor that it is free from the risk of infringing third-party rights.
A particular risk arises where AI tools are used by employees, external associates, marketing agencies or developers in projects for a client. If the contract does not clearly regulate whether the use of AI tools is permitted, who is responsible for the result and which rights are transferred to the client, a serious dispute may arise in relation to ownership, rights of use and liability.
For this reason, contracts increasingly regulate the use of AI tools in the creation of agreed work results, including materials, content, software solutions or other work products to be delivered to the client. This is particularly important in contracts for software development, marketing services, design, content creation, production, consultancy services and other projects in which the result of work has economic value and is intended for further commercial use.
Risk of Infringement of Third-Party Rights
One of the most significant practical risks is the possibility that AI-generated content may infringe third-party rights.
An AI tool may produce text, a photograph, illustration, design, musical work, software code or other content that is similar to an existing copyright work, trademark, industrial design, brand identity or other protected subject matter.
This does not mean that every AI output is legally problematic. However, for commercial use, particularly in areas such as branding, advertising, software, product design, packaging and public campaigns, it is advisable to conduct an additional legal and creative review before publication or use.
Particular caution is required when an AI tool is instructed to generate content “in the style of” a well-known artist, brand, campaign, author, photographer, designer or company. Such an approach may increase the risk of copyright infringement, trademark infringement, unfair competition or other claims by third parties.
In a business context, the risk is not limited to the question of whether the content is “new” or “different”. It is important to assess whether the final result may create an association, among consumers or business partners, with another person’s brand, copyright work, protected design or recognisable creative expression.
AI Tools, Confidentiality and Trade Secrets
Intellectual property is not the only area of risk. Companies often overlook the fact that the use of publicly available AI tools may inadvertently disclose confidential information.
Employees or associates may enter into an AI tool:
- draft contracts;
- business plans;
- technical documentation;
- source code;
- databases;
- client information;
- internal strategies;
- know-how;
- data constituting trade secrets.
If a company does not have a clear internal policy on the use of AI tools, there is a risk that confidential information may be processed in a manner inconsistent with contractual obligations, trade secret protection rules or personal data protection regulations.
It is particularly problematic when employees use publicly available AI tools without the employer’s knowledge, without a prior assessment of the applicable terms of use, and without clear rules on which information must not be entered into such systems. In such situations, the company may lose control over information that has business, technical or commercial value.
Therefore, the manner in which a company regulates and controls the use of AI tools is not only a technological issue, but also a legal and organisational one.
The EU AI Act and Copyright
Although the Republic of Serbia is not a member of the European Union, EU regulation is relevant for domestic companies that operate on the EU market, provide services to EU clients, use AI systems in cross-border business or develop products that may be made available to users in the EU.
The EU AI Act represents the first comprehensive legal framework for the regulation of artificial intelligence in the EU. The Act regulates risks associated with AI systems and establishes a regulatory framework for their use.
It is particularly important that providers of general-purpose AI models are subject to obligations relating to transparency and compliance with copyright law, including obligations concerning technical documentation and information on the content used for training the model. Within the Code of Practice for general-purpose AI models, transparency and copyright are specifically addressed as relevant obligations of providers of such models.
For companies using AI tools, this does not mean that they automatically become subject to all obligations applicable to providers of AI models. However, the EU legal framework indicates a clear regulatory direction: the use of AI tools must become more transparent, more responsible and more legally controlled.
For domestic companies, this means in practice that AI cannot be regarded as an unregulated area. Even where specific obligations under EU regulations do not apply directly, the standard of care in business, client expectations and contractual requirements are increasingly moving towards documented, verifiable and controlled use of AI systems.
What Should Companies Regulate Before Using AI Tools?
Companies that use AI in their day-to-day business should introduce basic rules and contractual mechanisms to mitigate risk.
In particular, it is advisable to regulate the following:
Internal Policy on the Use of AI Tools
It is necessary to define which AI tools may be used, for which purposes, under what conditions and subject to which levels of approval. The internal policy should be sufficiently clear for employees to understand when the use of AI tools is permitted and when it requires prior legal, technical or management review.
Prohibition on Entering Confidential Data into Public AI Tools
Employees and associates should clearly know which information must not be entered into AI systems. This particularly applies to confidential contracts, client data, trade secrets, code, technical documentation, financial data and internal business strategies.
Rules on Reviewing AI Output
Content intended for public or commercial use should not be published without human review, particularly where it relates to branding, advertising, software, design or legally sensitive materials. Human review should not be merely formal, but should involve an actual assessment of accuracy, originality, brand compliance and the risk of infringing third-party rights.
Contractual Clauses with Freelancers, Agencies and Developers
Contracts should contain provisions regulating whether the use of AI tools is permitted, under what conditions, who is liable for infringement of third-party rights and which rights are transferred to the client. Where appropriate, the contract should also provide for an obligation to notify the client of the use of AI tools, warranties regarding the originality and legality of the agreed work results, as well as liability for damages in the event of infringement of third-party rights.
Evidence of Human Contribution
Where it is important to prove an authorial contribution, it is useful to retain drafts, versions, instructions, modifications, creative decisions and other evidence showing that AI did not independently create the final result. Such evidence may be significant in the event of a dispute concerning authorship, ownership, originality or the scope of transferred rights.
Review of the Terms of Use of the Specific AI Tool
Different tools have different terms regarding rights in the output, commercial use, processing of input data and liability. Before using an AI tool for business purposes, particularly in client projects, it is necessary to verify whether the relevant terms of use permit the intended use of the result.
Particular Relevance for Marketing, IT and Creative Industries
Risks relating to AI and intellectual property are particularly pronounced in industries in which creative or technical work products are created on a daily basis.
For marketing agencies, the most sensitive issues are visuals, slogans, campaigns, copywriting and brand identity. AI may accelerate the creative process, but the final material intended for public use must be legally and commercially verifiable.
For IT companies, particular attention should be paid to AI-generated code, open-source components, licences, documentation and liability towards clients. The use of AI tools in software development may raise issues concerning ownership of code, licence compatibility, confidentiality and security.
For media and publishing, important issues include authorship, editorial control, sources, permitted use of third-party content and transparency.
For startups and SaaS companies, AI may form part of the product itself, which further raises questions of ownership over technology, data, models, output and liability towards users.
Conclusion
Artificial intelligence significantly accelerates business processes, but it does not eliminate the rules of intellectual property, contractual liability and protection of confidential information. On the contrary, the more AI tools are used for commercial purposes, the more important it becomes for companies to regulate in advance the legal framework for their use.
The key risks relate to whether AI-generated content may be protected by copyright, who has the right to use it, whether such content infringes third-party rights, whether confidential data is secure and who bears liability if a dispute arises.
For this reason, AI should not be viewed merely as an operational tool, but as part of a broader system of legal, contractual and compliance control. Companies using AI in their business should have clear internal rules, appropriate contractual clauses and a process for reviewing AI output before its public or commercial use.
Properly regulated use of AI tools may bring a significant business advantage. Unregulated use, however, may create risks relating to copyright, confidentiality, liability and the company’s reputation.