The talks of AI taking over the world is all around us copiously. Mention an industry and the discussions of AI taking over it takes over the conversation, at least. The Finance and Accounting Industry isn’t untouched by the inadvertent talks of AI taking over it.
To clear the air on all such talks and have an objective view of the impending metaphorical “AI-apocalypse”, we had our Head of Global Finance and Accounting Outsourcing Services, Datamatics Business Solutions, Inc., Hitendra R. Patil, sit down with Tara Shankar Jana, fondly known as TJ, the host of The is Not Another Bot: The Generative AI Show, podcast and give us his valuable insights on the entire topic. The podcast enthusiasts can head over to the link below and listen to the whole discussion on a podcast platform of their choice.
We have transcribed the entire discussion into an interview format for our beloved bibliophiles below. So, read on to learn the valuable insights Hitendra shared during his conversation with TJ.
The Real-World Capabilities of AI in FinTech
Tara:
What’s factual, what’s the hype, and what’s practically realistic about AI in FinTech today?
Hitendra:
Let’s start by addressing the elephant in the room, when we talk about AI in FinTech, we are not talking about the ChatGPTs and Google Bards of the world only. Those are the brightest stars in the galaxy that is AI, but that galaxy is way bigger than just those two stars.
With that being said, let us look at what’s factual when implementing AI in finance and accounting. We are already seeing the increasing use of Chatbots and automation for data analysis, real-time insights, and more. In some use cases, we have even seen the use of tools like ChatGPT. However, concerns have been raised about it, in terms of what’s within and outside the accounting world, by the accountants. And rightfully so.
On the “Hype” part of the question, I believe we are still very far from attaining “full automation.” The biggest hype we see is around full automation, with statements like everything will be automated, and accountants will lose jobs. I don’t think anything as drastic as that will happen anytime soon; the simple reason behind that being the fact that the data used to train these large language AI models is not the only thing that drives the accounting world. There are other aspects, such as the client requirements, the expectations, Etc. As an accountant, you need to merge these two worlds, the world of data and real-world expectations, which is a mammoth task, even for something as powerful as AI. Accountants make complex judgmental calls based upon various moving parts within databases.
Coming to the what’s practically the realistic part of the question, when it comes to being practical, I feel things like cleaning the data, aggregating data, validating the data, etc., all these things can be significantly improved using AI. Even AI-curated data seems within the bounds of possibility. One more area that seems highly practical is audit. With AI, it is now possible to audit large datasets without missing out on minor anomalies or frauds. Doing a 100% audit seems highly likely with AI, which means detecting frauds and preventing them from occurring altogether. In addition, there is What-If Analysis, Predictive Analysis, Etc. All those things can be run relatively efficiently using AI instead of running them on an Excel sheet like earlier.
Introduction to AI & First Impressions
Tara:
When was your first interaction wth modern AI and how it was?
Hitendra:
The first time I really learned of AI making inroads into mainstream world, was back in 2006 when the founders of Google talked about their “Search Engine,” which not only is capable of finding any information you are looking for, but also knows the context behind it. However, that was AI in its early days, but was good enough for me to get hooked onto it. I started reading about it and learning everything I could about AI. And finally, I had my first introduction with an AI tool in IBM Watson, 3 – 4 years after I first heard of AI. It was mesmerizing from the first interaction. I was enamored by its capabilities of decoding natural language query, sentiment analysis etc. As I learned more about natural language processing (NLP), AI became even more magic-like for me. We started leveraging the power in our organization by training it on thousands of natural language queries, and deploying it to shorten the customer query resolution time, and improve the overall customer support cycle.
Tara:
With that, let’s move towards the next segment of our discussion, which considering your background in customer centric roles, is very apt for you.
Generative AI & Customer-Centricity
Tara:
Could you share your thoughts on the impact of generative AI on customer experience and how it will change it?
Hitendra:
I am a keen student of Human Behavior Science, so when I saw that AI could also do sentiment analysis, that opened a new doorway of opportunities. In a normal workday, as an organization, you get thousands of emails and calls, making it harder for you to focus on any single one, consequently hampering the overall customer experience. However, with the power of Generative AI in the form of ChatBots, you can empower customers to find the answers to their queries independently. You train the generative AI models on thousands and thousands of natural language queries to make them efficient and accurate in addressing customer queries. Accounting firms can leverage AI to guide customers towards the best possible action to make their accounting work 100% compliant with the regulations.
Furthermore, Generative AI can also be used to reduce data handling time and allow accountants to invest their time and expertise wisely in resolving client queries. Using large language models, AI can help accountants improve their communication skills, bringing down the customer query resolution time and making the response more accurate and precise. It’s worth mentioning that organizations can try and bring in the ChatGPT-like feature in their tools and systems as well to make them more intuitive for their customers. However, when that will happen needs to be seen. However, I see software’s ability to predict the duration of a client’s business using AI and many such possibilities emerging in the future.
Tara:
In continuation to what you just said, because it’s all about improving customer experience, do you think companies should be using AI, or Generative AI per se? Also, what would you say is the biggest roadblock in adopting generation AI or AI in general.
Hitendra:
To begin with, tools like ChatGPT and Bard, as of now, are predominantly text-based. So, they can be positively and negatively influenced using text-based models. However, as accountants get familiar with these tools, they can extract the maximum out of them by inputting the right prompts. However, it will take time, and considering accounting is a time-sensitive practice, we see accountants generally giving less importance to these things. Also, if I have to highlight the biggest roadblock, I would like to bring your attention to the fact that accounting is pretty much regulated. Every change you bring into the space must comply with the authorities. However, the good thing is that regulatory authorities are making necessary amendments to accommodate these AI-driven changes. Another roadblock I see is the skill gap. AI, or generative AI for that matter, is still in its infancy; we need to work with it to acquire the necessary skill set to work with this revolutionary technology.
AI for Tackling Unethical Practices in Accounting
Tara:
How do you think leaders in the accounting space can leverage AI to handle unethical accounting practices and what do you think is the possibility of regulatory authorities working with AI to conduct such activities?
Hitendra:
Regulatory frameworks for accounting and tax are rather complex; you add cross border transaction to it, and it becomes even more complex. Fitting all those frameworks in one AI tool will be rather difficult. Add to that the numerous complex rules that apply to a business, and it becomes even more unconceivable to ensure that you are covering every single aspect of every rule that applies to a business. This is where the expertise of accountants comes into play. It is beyond the realm of AI. Add regulators into the mix, further elevating the situation’s complexity. However, that is not to say that AI as a preventive measure for unethical practices is impossible. We need many more evolved and advanced AI models to successfully tackle the accounting world’s complexities. Maybe 5 or 10 years down the line, we will have it.
Using AI to Drive Efficiency in FinTech
Tara:
What’s the balance you see in using AI to enhance the work of existing accountants? Also, can we or customers see a potential cost-cutting in the future?
Hitendra:
For me, accountants were never supposed to serve solely as data handlers. The whole idea behind handling data was to help the customers ease up the financial lives of the customers. However, now, with the arrival of AI, accountants can finally get their focus back on offering strategic services to their clients. Currently, the focus predominantly was on data handling, but that is all set to change now. The accountants will now be able to apply their expertise to client situations. Accountants can also expect new jobs to emerge in the market, something I had predicted 4 – 5 years ago. Accountants will now become AI Accountants, which is also the title of my book, “Rise of AiCCOUNTANTS.” We can expect pretty much every role in the accounting world to have a slight flavor of AI. The auditors will now be AI auditors, and so on. Overall, I feel that all the technologies we have developed augment human beings, and AI is no different. We must also treat AI as a means to increase the ability of human beings in their respective fields, and it will be.
Where Do You See AI in 10 Years’ Time?
Tara:
Where do you think AI will be ten years from now, considering we still are in the very early days of AI adoption?
Hitendra:
I wish to see AI extracting all my wisdom and putting that to work for me so that I can retire comfortably. On a serious note, though, full automation is the direction I see AI heading towards. Automation of repetitive, data-intensive tasks will continue to happen to a level where we can safely call it fully automated. Automated reporting is another avenue I see AI heading toward. There will also be a significant improvement in regulatory compliance; predictive analytics for auditing is another avenue likely to be influenced by the development of AI. The biggest thing for accountants will be business advisory and strategic consulting.
Closing Remarks
From the outset, the podcast gives deep insights into AI and Accounting and the coming together of the two. For CPAs and accounting professionals, the podcast is a great learning experience, especially if you are looking to strike a balance between the emergence of AI and its impact on the role of accountants as we know it.
In his closing remark, Hitendra wisely advises enterprises and young accountants to embrace technology instead of dreading it. A good starting point is to identify the gaps that technology fails to address in its first iteration; that is where you will find your market. Rest everything is an enabler. If you can figure this out, you are set for a fruitful career, irrespective of the changes or developments in the technological world.
For more such insights and thought leadership, you can reach out to Hitendra at [email protected].
