Archana Arakkal, Machine Learning Engineer at Synthesis
The purpose of digital transformation is to allow an industry to perform at an optimal pace, as well as to leverage newer technologies out there. This results in a more effective working environment.
If we look at the banking sector as an example, digital transformation has assisted the industry to transform by using new technologies. The adoption of new technologies helps change the way in which operational structures work.
As an example in the banking sector, a customer applying for a loan would physically have to go into their financial services provider; manually fill out forms that would get filed away in a cabinet, and the branch would take 2-3 months to process the application. Enter Artificial Intelligence (AI). Now using new technology it can be done in seconds. Or milliseconds.
Object Character Recognition (OCR)
An interesting development in the AI space is Object Character Recognition (OCR). In the not so distant past, a person applying for a loan could scan the form; now with OCR, a photograph on a mobile phone is good enough. Thanks to machines that recognise objects and characters and extract the content, transform it into plain text, there is no longer any need to go into the institution.
With AI we are becoming smarter with how we work by optimising the way businesses function, and we are doing this is less time too.
Machine Learning can propel digital transformation
For some, AI is a buzz word and there certainly is a lot of fear around it, but the reality is, it can propel organisations forward, especially in areas they are struggling with. For example, fraud detection and customer profiling. The benefits and advantages specifically in these two areas are obvious when the role of machine learning is clearly understood. Fraud can be detected quickly, and in the case of customer profiling – done accurately there are many ways businesses can use this information to reach the individuals who make up their customer base.
Partner with an Insurer you can trust
In an uncertain financial climate, our client-centric approach of developing strong relationships with partners and clients, while boasting a deep understanding of their business, helps us to create unique solutions.
Centriq’s insurance subsidiaries are authorised financial services providers
Clean data and data-driven decision making
Using machine learning and AI depend on the foundation of the company’s data. Like everything – if data is accurate, the results are excellent. However, organisations do have legacy issues, and data can be inaccurate and misconstrued.
This is a challenge as businesses need to fix their data before they can reap the enormous benefits of clean data. Here, interestingly enough, humans are needed to step in and correct information. Humans will always be needed, but we need to get to a place where we use technology to our advantage. AI can help to get the data to the place it needs to get to.
Let’s start using AI to augment humans so we can fix what exists. It can’t be one or the other – we need a combination of AI and human interaction.
Why AI matters
It’s not only machines getting smarter, but people are also becoming more tech savvy and more demanding. In order to keep up, we need to start leveraging technology and infrastructure. Not only are we going to be using AI to enforce digital transformation, but we are catering for a market that demands it. AI is not something in the future – AI is already here, and if your organisation is not already using it, you are falling behind!
We are a bunch of numbers
Gone are the days where one person has one footprint. Today, the average person has three devices on which they are interacting. On average this one person has about five accounts, such as social media, banking apps, fitness apps, shopping apps etc. This means that there are more digital footprints than there are existing human footprints.
Bot nor not
In South Africa, we are reaching a point where AI has matured to the point where human interaction is no longer necessary in some instances. For example, Google has released an AI model Generalised Pre-Transformed Trainer (GPT3) that can understand language and talk back to you (in a call centre environment), and you will not be able to differentiate whether it is a bot or a human!
In conclusion, the fear around AI is null and void; if you look at the good it definitely outweighs the bad – AI can be so useful. Yes, fear is driven by the unknown, but the reality is, that people will always be needed.