Year: 2024
Type: article
Abstract: With the development of artificial intelligence (AI) systems, their popularity is growing. Such systems are used in many areas, from customer analytics and search engines to voice assistants and medical research. The tasks assigned to the systems are becoming more and more complex, and therefore artificial intelligence often needs to operate on confidential data; the results of the system's operation can have large-scale consequences. This creates a new problem: the problem of trusted artificial intelligence. The authors set the goal of systematizing knowledge about possible threats and vulnerabilities associated with the use of AI technologies, analyzing existing standards in this area, and also identifying and describing relevant technologies that can increase confidence in the use of intelligent systems.
Source: INFORMACIONNYE TEHNOLOGII
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Keywords: AI security, AI risks, AI vulnerabilities, AI standards
Open Access status: bronze
Funders: Australian Government, Office of Science
OpenAlex ID: W4401547013
