A bank has hired a “ robot lawyer” that has pored through millions of documents and works 30 times faster than humans.
Created by the company’s legal department, it analyses legal information, verifies whether businesses are bankrupt, and decides whether loans can be issued to potential customers.
Alexander Vedyakhin, first deputy chairman of Sberbank’s executive board, told Computer Weekly : “Each document is translated from scanned copies into a machine-readable format and the text of the document and other significant fragments are extracted, for example, stamps and seals.”
Another tool then goes over complex facts from the documents using machine learning.
The process is made significantly quicker by the technology – a human would typically take 30 minutes to go through one of the documents, but the robot requires less than a minute.
Mr Vedyakhin continued: “Traditionally, a lawyer manually checks relevant information in a package of documents and makes a decision independently, guided by accepted rules and their own experience.
“The result of this activity is a conclusion, which takes into account all possible legal risks.
“To automate this stage, a ‘digital lawyer’ implements a decision-making system that prepares a legal opinion based on the knowledge of lawyers, laws and internal regulations.
“For example, the program can verify that a CEO, acting on behalf of the client, is really entitled to obtain a loan from Sberbank.”
The company previously earlier deployed customer support robots – Promobots –that carry out initial consultations and facial recognition.
Earlier this week, scientists unveiled an AI model that can detect is someone has coronavirus purely from the sound of their cough.
The AI algorithm, built in the US, correctly identified patients who were infected even if they had no other symptoms.
Researchers say humans are unable to hear the vital difference in the sound of someone with a cough who is asymptomatic.
They say this is because Covid-19 changes the way you produce sound, even if there are no symptoms.
The model, developed by the Massachusetts Institute of Technology (MIT) in the US, was 98.5% successful in detecting people who had officially tested positive for Covid-19.
And it could tell an impressive 100% of cases of people who had no other symptoms.