Artificial Intelligence technology offers computational, decision-making, and optimizing abilities that surpass every previously established traditional computation method. By being able to navigate across large amounts of data, the realized solutions learn on their own and provide results that would be unattainable with other ways. The complex nature of the AI makes the solutions difficult to define. The potential is left untapped, as the solution developers struggle to define product features that would correlate with the user needs. On the other hand, the the potential AI solution users may not be aware of the AI capabilities, possibilities of integration, and automatization of the operation processes.This creates situation, in which the potential of the AI is not used effectively, and areas that could benefit from it are not able to integrate AI systems in their operation. This research looks at this issue from the perspective of the value proposition design process. This process assumes a definition of potential benefits and uses on the side of the developer, aligning, and communicating them with the potential users. Specifically, the Value Proposition Creation model is used, as it is a method often used by AI startups and university spinoffs. The main question this research asks is “What are the factors influencing the specific value proposition of innovative AI solutions”. By answering this question, the paper hopes to establish a method for understanding the specific aspects of the AI solutions that should be taken into consideration while designing an effective value proposition and communicating it with the potential user. To answer the research question, an in-depth look is taken at the Artificial Intelligence adoption processes. First, a literature study helps define the factors involved in the AI adoption. A conceptual model is created, which is then evaluated via a series of semi-structured interviews with AI research and development experts. By doing so, the relationships between the factors are obtained and a general impact on the adoption process is understood. This allows for a formulation of the relationship of said factors with the Value Proposition Canvas. Additionally, literature research is conducted on the value proposition frameworks, which allows for a definition of good practices in value proposition design in relation to the Artificial Intelligence solutions. The obtained frameworks are then tested in a single exploratory case study, which looks specifically at the idea of Deep Reinforcement Learning algorithms used as a decision-making method in the context of road maintenance planning. The research provides an overlay on the Value Proposition Canvas, obtained through the evaluation of the AI adoption process. Such approach creates a framework, which then can be used to effectively define and clarify the individual values, benefits, gains, and features of an AI product. Moreover, by providing the interrelationships between the adoption factors, understanding of the internal dependencies of factors necessary in the product development process can be obtained. The framework can be used by developers in the field of Artificial Intelligence to assess the necessary requirements of the solution, highlight the key areas that have to be researched, as well as help in communicating of the crucial solution aspects with the potential clients and users. The framework can be best used in the context of an AI startup or a university spinoff, because of its generalist approach. It assumes a fast development of a value proposition, which is necessary in the context of a Minimum Viable Product definition. The results of the research combine the adoption processes of the AI solution and the method for their value proposition, realized with the VPC. It has been shown that the adoption factors may play a relevant role in the value definition of the potential solutions. Because the adoption models may not be used directly in the VP processes, a framework of questions has been setup for the solution developers to reflect upon during the Value Proposition Creation. Such approach is in line with the general methodology of the VPC, as it uses questions to define the relevant fields of the canvas. The goal of the research was not to modify the canvas itself, as its intrinsic agility and simplicity is the core strength. Moreover, by looking at the common practices seen in other Value Proposition frameworks, good practices have been defined for the VPC in relation specifically to the Artificial Intelligence solutions. These results have been afterwards validated in the exploratory case study for the Deep Reinforcement Leaning decision-making system in the field of road maintenance, indicating the specific steps that the solution developer must undertake and creating an example Value Proposition Canvas for this specific case.