Artificial intelligence (AI) tools such as ChatGPT and Microsoft Copilot are widely employed in South African workplaces for tasks such as email composition, document summarization, document writing, meeting minutes, statistical analysis, and information research.
As such, AI is being integrated into everyday business practices, from supporting recruitment through AI-enabled applicant screening and automatic job description generation, to facilitating customer interactions through chatbots and virtual assistants, and providing analytical support.
The early use of AI in the workplace was informal in many workplaces across South Africa; There is a growing trend to establish work policies governing the use of AI in the workplace. Businesses are now establishing AI strategies aimed at addressing data risk, compliance, and productivity alignment. Statistics show that the use of generative AI (GenAI) in businesses is rapidly increasing as a formal workplace tool rather than informal, unregulated use.
In fact, statistics show that South Africa’s AI adoption rate is high and generally above the global average, and GenAI systems/tools are becoming mainstream in various sectors. According to the survey, 75% of respondents Think about how AI can help you in the workplace. More job postings are also requiring AI skills as part of their required skill set.
The rapid increase in the use of AI in South Africa’s workplaces poses a significant risk of copyright infringement, which can lead to liability for employers. The GenAI system interacts with the user through: “promptIn the context of a GenAI system, a prompt is an instruction or query by the user that is intended to guide the output. Prompts serve as the primary language for communicating with the GenAI system.
Prompts come in a variety of formats and vary in complexity. These include simple queries, instruction-based prompts, role-based prompts, context and task prompts, and continuous refinement prompts. Prompts may also reflect the types of use of AI in the workplace described above.
The rapid increase in the use of AI in South Africa’s workplaces poses a significant risk of copyright infringement, which can lead to liability for employers.
Personal use of AI in the workplace can create copyright risks and infringements in a variety of ways. Through prompts, employees can May input copyrighted material into the GenAI system. This can be done by: directly copying and pasting protected text, requesting transformations based on existing works such as paraphrasing or summarizing paywalled academic papers, or uploading proprietary or licensed material such as corporate contracts or confidential manuscripts.
Retention or further processing of this material by the Platform beyond temporary use may pose significant risks from a copyright and licensing perspective. In countries that recognize copyright, including those that have ratified the Berne Convention for the Protection of Literary and Artistic Works, The copyright owner has exclusive rights to reproduce, adapt and distribute the material..
When an employee inputs copyrighted material into a GenAI system, questions arise as to whether that activity constitutes reproduction and whether the output constitutes a derivative work. If the answer to the above questions is positive, copyright infringement concerns arise.
The question then arises as to how an employee’s personal use of AI may lead to employer liability. At the current stage of the development of AI in the workplace, AI does not replace human employees, but works with them to complete work-related tasks. It is important to note that employers who use AI systems within the workplace are entitled to: Use such systems professionally.
The increased use of AI systems in the workplace is correlated with increased harm resulting from unethical or dangerous use of such systems. Using IT-vetted AI tools in the workplace carries legal risks that can lead to employer liability. Employers are exposed to significant risks of copyright infringement and subsequent liability, as employees may feed copyrighted content to AI systems to complete work-related tasks, shaping the training and knowledge of such AI systems. Essentially, any unauthorized use of intellectual property by an employee through prompting exposes the employer to legal claims.
It is important for employers to prevent risks arising from the use of AI systems in the workplace by their employees. Employers are encouraged to determine the extent to which AI will be used in the workplace, train employees on restrictions on AI use regarding copyrighted content, ensure AI output is reviewed to ensure it does not contain copyrighted content, and ensure employees understand the legal risks of using AI.
The above recommendations are important in protecting employers from legal claims arising from intellectual property laws and copyright infringement.
• Ndlovu is a chartered lawyer and researcher whose research focuses on the intersection of law, technology, and human rights.
