Article continues below
Want to learn more? Check out some of our courses:
Episode highlights:
- Artificial intelligence can be used to improve the banking industry, but there are ethical guidelines, principles, and regulations that should be followed to ensure that these decision-making tools work for everyone equally.
- To be able to use AI technology to their advantage, many banks choose to improve their current legacy systems and infrastructure by developing a completely different system rather than updating their current one to improve their effectiveness.
- Up-skilling employees is a must, and helps ensure that staff is capable of improving their own processes and using the company’s systems more efficiently.
Transcript (edited for clarity)
Claudia: What is the last thing you learned that got you excited and why?
Marino: Since the pandemic started, two things caught my interest. I work in cybersecurity from a governance and policy perspective, but I do not code. I wanted to learn more about the infrastructure, hardware, and hacking, the real skills people need from the engineering side of things, so I enrolled in a cybersecurity program. It was something that I wanted to learn not to become an expert, but just to understand how it worked so I can do my job better. And the second thing is something I am currently working on; I want to start writing blogs and opinions based on my experience in the field of risk management. Writing is something I want to improve because I am passionate about it. Hopefully, I will start publishing things soon.
Claudia: What do you think is missing in the banking industry in terms of banks’ ability to use data effectively?
Marino: Three things are missing. First, the legacy systems, the legacy infrastructure, and the technology used nowadays are outdated. We are still using systems and infrastructure that we have been using 40 years ago. People are still using spreadsheets and manual inputs for data entry. I think you need to move all the existing broken processes before you can implement automated processes, and then start using artificial intelligence to make decisions. The second is - I do not think banks have the right skill sets now. Banks need their own professionals, artificial intelligence experts, data analysts, and engineers to implement AI, as they are currently competing with Amazon, Apple, the technology firms. The banks are not able to have the right skill sets, senior management, and the right teams to implement artificial intelligence in the actual system. They also need to collaborate with other banks when it comes to implementing certain methods and aspects. The banks possess an incredible amount of data. If they do not start collaborating on certain aspects then the customer, which the bank effectively exists for, is not going to benefit. That is the reason why the senior management of every bank needs to invest in improving that technology and systems to a stake, not that it would be very valid right now but also for the next 10 to 20 years.
C: Imagine AI skills being in the toolkit of everyone in the banking industry. What do you think is the best way to get there?
M: The first element is up-skilling your current staff to understand and improve their work, whether we are talking about customer care, risk management, or finance. The up-skilling process needs to take place to make the staff more digitally capable of improving their own processes, but also to improve working with systems, and not being too scared of changing the way they work. The second element is the bank's legacy systems and infrastructure, and I can compare that with having an old train. Do you want to rebuild the old train with a new system, or do you want to build a brand-new train? Sometimes, it would be easier to build a brand-new train because it is more expensive to build the onboarding right, to rebuild it. This is the reason why a lot of banks are starting to build their own digital bank, which is similar to a side project. They are starting to plug the old infrastructure into the new infrastructure. And over time, the side project becomes the new infrastructure and system that people will start using in the next 10 to 20 years. I think that is probably the best way of pivoting and getting to where you need to be to use AI faster than trying to rebuild something that already exists, which is quite difficult and long. And I think that's why AI is taking a while to come into banking.
C: How has your work changed in the last year?
M: Before the pandemic, we used to work from the office for a hundred percent of the time, since we have a lot of regulations that require us to be in the office. We worked with our technology teams to bring everyone to remote working, which included getting everyone set up at home, getting the monitor set up, getting the input, VPN internet, all that set up in just two weeks. After six months of remote working, I got a bit tired, working from home was starting to burn out. Because you are home, you do not learn how to switch off. I have had to learn to rebalance myself, which I have probably done better in the last four or five months than the first 12. I'm glad I can go back to the office voluntarily because it's a good set of scenery.
C: What is a cool project or initiative that you are working on right now and that you want to share with the world?
M: I started doing a lot of writing and blogging around ethics, AI bias, and building my own website called Futuristic Risk, which is going to talk about how to effectively manage risk in this new environment. All my opinions are there, nothing from my firm, so I should probably mention that. I hope to publish that website in the next couple of months. I am involved in a couple of cool projects of not-for-profit organizations that look at artificial intelligence, cyber bias, ethics, providing submissions and reviews on things coming up for government, but other aspects as well. Hopefully, we will continue developing and I will be in a position to publish it in maybe four or five months. It's something I'm definitely excited about.
