In recent years, artificial intelligence technology has rapidly penetrated into all walks of life, and the financial industry is no exception.
Many financial institutions have begun to try to apply artificial intelligence technology to the field of risk prevention and control, using technological innovation to prevent financial risks.
Currently, our country is actively trying and exploring in the field of "artificial intelligence + risk control" and has a certain first-mover advantage compared with its peers in the international financial industry.
At the World Artificial Intelligence Conference in July 2023, Tencent released a large financial risk control model.
In November of the same year, Tencent jointly formulated the world's first major financial risk control system with scientific research institutions and financial institutions such as China Academy of Information and Communications Technology, University of Science and Technology of China, Nanyang Technological University in Singapore, Zhongyuan Consumer Finance, and WeBank. Model international standard.
What will artificial intelligence technology bring to financial risk control?
In theory, artificial intelligence empowers risk control, reduces human errors and interference, and can improve the efficiency and accuracy of risk identification.
However, considering that artificial intelligence technology is a new thing that is still developing and is still immature, hasty promotion in the field of financial risk control may bring new risks.
Of greatest concern is the risk of data breaches.
Currently, many financial institutions choose to cooperate with technology companies with artificial intelligence technology in the field of risk control, and these cooperation often involve data sharing.
Large artificial intelligence models rely on a large amount of sample data for training. The scale and quality of the data have a crucial impact on the accuracy of risk control.
Theoretically, the richer the data, the stronger the ability of large models to accurately profile users, and the higher the accuracy of identifying risks in aspects such as credit approval.
However, as more and more data is shared, whether privacy can be effectively protected has become a new risk challenge.
It is worth emphasizing that financial data not only has the general characteristics of data, but also contains important content such as national account information and corporate capital flows. This means that once financial data is leaked, it may bring greater risks than general data leaks.
In addition to data breaches, legal risks cannot be ignored.
Historically, revisions to laws and regulations have often lagged behind the application of new technologies.
At present, artificial intelligence technology still has the possibility of generating false content due to errors in data and algorithms, and may cause user discrimination to a certain extent.
Once a large model generates inaccurate financial risk control reports, it will be difficult to distinguish whether the technology provided by technology companies is unreliable or the data provided by financial institutions is unreliable. This makes it difficult to define legal responsibilities, and it is easy for financial institutions and technology companies to The phenomenon of companies passing the blame to each other.
During the wrangling process, customers' reasonable demands such as loan approval may be delayed, and the risk will ultimately be borne by the customer.
The Central Financial Work Conference proposed that financial supervision should be comprehensively strengthened to effectively prevent and resolve financial risks.
In view of the new risks that may be brought about by the application of artificial intelligence technology in the financial industry, on the one hand, we must improve laws and regulations to protect people's right to reasonably question the results generated by artificial intelligence technology, and ensure that artificial intelligence technology is subject to accountability mechanisms and is transparent, fair, and safe. On the other hand, it is necessary to effectively manage financial data information, steadily and prudently promote the application of artificial intelligence technology, continuously improve risk control technology, improve and optimize risk management and prediction models, make technology better, and prevent artificial intelligence. Potential risks brought about by the application of intelligent technology in the financial field.
Su Ruiqi (Source: Economic Daily)