Examining the legal challenges and loopholes of artificial intelligence by considering the upstream documents of the country

In this article, the challenges and loopholes of laws in the world have been examined first. Issues that are generally challenging in the field of artificial intelligence and require risk management have been investigated. Based on this, some of the most important challenges and risks include the lack of algorithmic transparency - the lack of the possibility of protest - unfairness, bias and discrimination - liability for damages, etc. Also, due to the importance of artificial intelligence, the challenges related to this issue have also been investigated and studied. And finally, for the policies extracted from the previous phase of the project, legal criteria have been determined to evaluate these policies. Each of these legal criteria has been adapted to the obtained policies based on legal dimensions and aspects. The purpose of this process was to determine the legal loopholes in the above-mentioned documents based on the subject of artificial intelligence. For improvement and high accuracy, as well as for better analysis on the criteria that were adapted to the extracted policies, the data has been aggregated in an Excel file. For this purpose, the multiplicity of legal criteria in each of the legal dimensions related to each document was specified. Also, at the end, each of the documents is weighted based on the importance of connection with technology and the up-to-date ness of the subject in relation to the field of artificial intelligence. Based on the obtained analysis, the legal gaps in the documents related to artificial intelligence have been identified. Therefore, according to the rapid process of artificial intelligence, legal loopholes in the formulation of laws should be considered more sensitively.