After discussing the function that artificial intelligence plays in AIG, the CRO of AIG Fabrice Brossart said that it ought to be better referred to as artificially inflated.
He is convinced that there’s no adequate penetration into different sectors of artificial intelligence (AI).
The former Lloyds Group developer and Secco’s current CEO Chris Gledhill also shares a similar opinion as far as AI is concerned.
Chris Gledhill is most impressed with neural networks, which is a sub-sector of artificial intelligence (AI).
Artificial Neural Network (ANN) is a representation of a human brain’s biological neural networks.
Aside from comprising interconnected processes, ANN creates algorithms that can be utilized in modeling complex patterns and, in turn, helps the user in the decision-making process.
READ MORE: 10 Applications of Machine Learning in Finance
READ MORE: ANZ Bank Leverages Neural Networks to Avoid Dangers of Deep Learning
Gledhill told bobsguide that neural network AIs would mostly affect the back office, particularly for functions that are performing unskilled tasks, which are mostly outsourced to offices based in India.
According to him, such roles will be automated. Chris Gledhill suggested four ways relating to how neural networks will disrupt banking as follows:
1. Enhanced Evaluation of Loan Applications
For banks to grant a loan application, their aim is mainly to minimize the rate of failure of loan applications.
Doing so, in turn, increases the returns from the loans issued. As such, more banking institutions are currently using ANN to help them in making decisions for granting loan applications.
This process works not only through evaluating past failures but also by making decisions based on previous experiences.
2. Assist Credit Card Companies in Finding the Ideal Customer
Credit card entities require finding the ideal customer in a bid to maintain sustainable revenue.
In case the customer does not make enough use of the card, the profit is affected considerably.
In such a case, the incidental and incremental expenses would surpass the revenue, which would then lead to a non-profitable enterprise.
Due to this, additional credit card companies have embarked on leveraging ANN in an attempt to spot a suitable client that would produce enough revenue, principally through the analysis of the customer’s credit-card habits.
MORE – RPA – 10 Powerful Examples in Enterprise
MORE – Top 25 AI Software for the Banking Industry
3. Stock Market Index or Forecast
ANN comes in handy when it comes to dealing with forecasting stock values and stock market indices.
They undertake such tasks through prediction based on various parameters and the use of past data.
ANN enhances the accuracy of the predictions by using more trained variables and hidden layers.
4. Better Fraud Detection through Image and Character Recognition
The capability of ANNs to take in various inputs, the process then in an effort of deriving the hidden and interwoven connections allows them to be useful in character and image recognition.
As such, more banking institutions can leverage ANN in bettering their fraud detection measures.
Does New Technology translate to New Ethics?
ANNs’ moral issues can be challenging. These neural networks are susceptible to creating their biases when it comes to identifying trends in data and learning patterns.
Such biases can, as a result, become prejudices, especially when the discriminated data parts correlate with the already defined segments amongst clients.
Gledhill recommends that ANN should be perceived as an important tool for human users as opposed to a be-all solution.
Source Bobsguide
