Využití autonomních AI agentů v nákupních procesech B2B a jejich dopad na výběr dodavatelů

The bachelor thesis examines the use of artificial intelligence in procurement processes, with a focus on autonomous AI agents in supplier negotiations and supplier selection. The aim of the thesis is, based on available case studies and professional sources, to map which specific activities these agents take over in procurement, which decision making criteria and data they use, and what benefits and limitations their deployment brings in practice. The theoretical part defines the role of the purchasing function in a company, describes current digitalization trends, explains the difference between traditional automation and artificial intelligence, and introduces the concept of autonomous AI agents in procurement. The methodological part summarizes the use of a qualitative approach, document analysis, and a multiple case study based exclusively on secondary data. The practical part focuses on five selected cases of AI agent deployment (Pactum at Walmart and Maersk, Ivalua, SAP Ariba, and Zycus MERLIN ANA) and describes in detail how they operate in tail spend negotiations, sourcing, and the orchestration of procurement workflows. The results show that autonomous AI agents can take over a significant share of routine negotiation and sourcing tasks, shorten negotiation and sourcing cycle times, generate measurable savings especially for smaller contracts, and free up capacity in procurement teams, while at the same time requiring high quality data, clearly defined boundaries, and realistic expectations regarding the reported benefits.