The Use of AI in Accounting - Here to Stay, or Just a Fad?
Recently, there has been a surge in the use of AI in both everyday life and within businesses. While it could be argued that it’s making our lives easier and helping to automate tasks, there is also an argument that we’re already becoming too reliant on it by using it to think for us, rather than to enhance ideas we already have.
So, how is AI used in accounting, and should we view this as a positive development?
1. Increased accuracy
Imagine a world where you could upload large volumes of data and obtain accurate results to any queries performed on it… This is the new reality of using AI.
It can go further than this, because not only can you obtain accurate answers to your queries, but AI can also flag any inconsistent inputs. Meaning reduced time spent hunting down pesky errors in the original data.
However, you need to keep in mind that the results given to you by AI are only as good as the query posed to it. Meaning, it could give you erroneous results if you give it misworded queries or attempt to make them too complex.
2. Enhanced decision-making
Data-driven decision-making is the backbone of all businesses. However, acquiring the data and analysing it can take time, especially if it’s a deep dive into historical client data.
Not only do you have to analyse the historical data, but you also have to combine it with forward-looking projections so that it provides an overall picture that is useful to decision-makers.
Using AI to analyse the data and combine it in a way that provides valuable insights can save time and effort for everyone involved. However, there are also drawbacks to relying solely on AI for decision-making.
Over-reliance on AI has led business owners to use it to make decisions for them; after all, it has all the necessary data to make informed decisions and is utilising a prompt designed to provide the answer they need.
However, that’s precisely the problem: the prompt you use can lead the AI to provide only the insights you’re hoping for. In psychology, this is known as a Confirmation Bias, where you specifically focus on research outcomes that confirm your way of thinking.
This can be a little complex to understand at the beginning, so let’s take the example of “Model the financial impact of launching [new product], such as entering new markets, or major capital investments.”. The prompt here is obviously something that the business has worked on, and they’re now interested in understanding what could happen if they launch. However, they’re focusing solely on the positive outcomes, such as entering new markets and new investment opportunities. Whereas, for a well-rounded response, there should also be mentions of any drawbacks, such as the cost of production and increased workload that could lead to staff burnout and turnover, including hiring costs.
Therefore, if the prompt were to be used as it currently is written, it would affirm to the decision maker that the new product launch would be a good move for the business, confirming what they already suspect.
3. Saving time
Increased efficiency, fewer human errors, and automated tasks make it appear as though AI is the ultimate solution for saving time in accounting. This saved time can then be spent in other areas of your business, such as product development or marketing.
However, the answers AI gives you shouldn’t always be taken at face value. Any analysis carried out on the data should be reviewed and sense-checked by a human, and any prompts used should be examined for potential biases.
So, is AI a fad in accounting?
Simply, it depends on how you use it and how much faith you have in the results it produces for you. If you’re leveraging it to save time in data entry and analysis, you might find that it works well for you. However, if you’re using it to aid your decision-making, you might find it falls short and only confirms what you’re already expecting.