Sber Credit business based on AI

Main articles:

• Artificial intelligence in banks
• Machine Learning

2020: Launch of AI-based business lending

On December 3, 2020, Sber announced the launch of an AI-based process for lending business with revenue from 400 million rubles.

Now the deadline for deciding on a large transaction will be less than two days. This process will be available to medium and large business segments with annual revenues of 400 million rubles. This was announced during a press conference by the vice president of SberBank, director of the division "Credit Products and Processes" Sergey Bessonov.

On December 14, we launch a pilot project for issuing loans under this scheme in the Central Russian and South-West territorial banks of Sberbank. Technology based on artificial intelligence will be able to use both current and new bank customers. We already offer our customers a universal boxed product - pre-approved ʺkredity for 7 minutʺ. Our dream process is a solution for providing financing on customer-customized terms. Already in 2021, we will replicate our dream process to all regions of the country, as well as expand its functionality: we will add revolving lending transactions with property security and state programs of the Ministry of Economic Development and the Ministry of Agriculture, provide automatic updating of the Cash Flow model, increase the variability of repayment schedules - Sergey Bessonov, Vice President of Sberbank, Director of the Credit Products Division

Features of the credit process:

• complete digitalization - from filing an application to receiving money for an account;
• The period before making a decision on the transaction is less than 2 days;
• Maximum castomization for requirements.

"Under the hood" of a seamless process for the client - a whole cascade of machine learning models using information from external and internal sources, which allows:

• Create a digital customer profile
• Create pre-approved quotations with transaction structure and limit amount that take into account customer needs, including covenant package;
• Automatically make decisions about transactions.