Exchange of information is an essential process carried out by tax authorities in the fight against cross-border tax evasion and ensure fair tax practices among taxpayers. To avoid paying taxes, taxpayers have taken advantage of the differences in tax systems of other countries. So-called "tax havens" are characterized by a lack of transparency, secrecy of banking data, and the refusal to share information with tax authorities of other countries. Information exchange helps fight tax evasion and promotes transparency in the global financial system. It enables tax authorities to obtain information they would not otherwise have access to. To put an end to the outflow of income to tax havens, the US government decided to develop a law aimed at combating tax evasion through the mutual exchange of information between tax authorities of different countries. FATCA (Foreign Account Tax Compliance Act) has led to the global development of automatic information exchange, particularly in relation to the exchange of information on financial accounts between tax authorities. This paper discusses the benefits of exchanging information on tax matters, as well as the challenges posed by the implementation of FATCA. After an introductory section, the second part deals with FATCA in general, its objectives, and the obligations arising from FATCA agreements. The third part describes different models of agreements and differences between information exchange models. The fourth part focuses on the implementation of FATCA in the Republic of Croatia, which became a signatory to the agreement in 2014. The Ministry of Finance of the Republic of Croatia played an important role in implementing the Agreement on Improving International Tax Compliance and implementing the US Foreign Account Tax Compliance Act by collecting information. The issue of tax fraud and tax avoidance is being addressed at the EU level, and the fifth part of the paper covers EU regulations. Additionally, the last chapter is dedicated to comparing FATCA with the alternative Swiss Rubik model. The final part of the paper presents the conclusions.