| Abstract [eng] |
This paper addresses the problem of allocating copyright of the images created by Artificial Intelligence (AI) as co-author. The rapid development of AI technologies and their integration in the creative industries is challenging traditional notions of authorship and property rights. A key question is how property rights are allocated when an AI acts as co-author in the image creation industry. The literature analysis has shown that current legal frameworks are not fully adapted to the involvement of AI in the creative process. There is considerable controversy about the role of AI in the creative process, which makes it very difficult to determine who should own the property rights in the solutions created by the AI. So far, there is a great deal of uncertainty in current law as to the copyright and ownership of works created with the help of an AI. When analyzing the literature on the application of AI in the creative process, it has been noted that there are a number of authors who argue that AI can be involved in various stages of the process, from the generation of ideas through to final refinement of the work. Also, it has been analyzed that Generative AI models can produce original results, sometimes even reaching human-level creativity. Based on the models proposed by other authors, explored in the literature analysis and using the insights and parts of their models, a conceptual a model for allocating copyright of images created by artificial intelligence as a co–author was developed. This model consists of two main parts - the copyright of human co-creator and the other part the AI as co-author. The latter is broken down into the main parts of the AI - the AI creators and the data providers, with the addition of a further part - the public goods management organization. The copyrights of the AI co-author would be distributed proportionally between these three parts, thus eliminating the need to create a separate legal entity for the IoT and distributing the property rights according to the existing systems. To test and refine the model, a qualitative study was carried out using semi-structured interviews with experts in AI and intellectual property. The results of the study revealed important insights for improving the model. Based on the experts' recommendations, the model has been modified: removing the section on creators of AI, as they are not directly involved in the creation of the outcome; introducing a section on “Collective Management Organizations” to protect authors' and performers' rights; relocating the data providers to this section; and adding a section on “Reinvestment in the creative sector” to provide a fund to be funded by the profits from the works created by AI. |
