Save content
Have you found this content useful? Use the button above to save it to your profile.
Weighing up the options when it comes to AI
istock_ArtemisDiana_AW

Where can AI make the biggest impact on accounting?

by

While accountants have taken a cautious approach to adopting artificial intelligence tools, the technology could have a major impact on day-to-day processes and operations, writes Tax Systems’ Russell Gammon.

1st Jul 2024
Save content
Have you found this content useful? Use the button above to save it to your profile.

With the dust now settling on the new wave of artificial intelligence (AI) tools, many businesses have elected to take a slightly more cautious approach to adoption. Rather than jumping in headfirst, firms are looking to identify key areas and use cases where AI can have a major impact before investing too heavily in the technology.

As I hinted at in a previous column, the labyrinthine and constantly evolving nature of accounting work makes the reality of AI adoption a complex subject.

However, there are key areas where AI-powered technologies could have a major impact on day-to-day processes and operations.

Simplifying trial balance mapping and data management

One area where AI is already advancing is the simplification of tricky but necessary procedures such as trial balance mapping.

Trial balance mapping, a “right of passage” for many juniors, involves taking the figures from a list of balances and transposing them into the right place in a tax return, as well as determining the right tax treatment of that balance. This time-consuming process involves lots of manual copying and pasting, inevitably leading to inaccuracies and human errors.

New AI-powered applications promise to make light work of such cumbersome processes. Not only can they effortlessly populate profit and loss (P&L) statements to an accuracy level of more than 90%, but by harnessing the power of generative AI, they can also prepare a tax treatment analysis to an accuracy level of more than 80%.

This enables tax professionals to identify anomalies much earlier in the process and deliver high-quality, accurate, consistent outputs. Leading solutions can even incorporate an organisation’s specific business context to further reduce the risk of human error.

Getting value out of tax data

At the process level, tax professionals can use AI-powered tools to create accurate reports, quickly perform calculations and file returns promptly. Rather than the traditional approach of having a plethora of pre-written reports available, it is possible to ask AI any question and get an answer back immediately.

For example, you could ask it to “draw a graph of the effective tax rates across my German entities” and it can perform that analysis in seconds. In general, we are seeing advancements in tax come down to one of two areas: either providing additional insight into the data or increasing efficiency in existing processes.

On the calculation front, it is important to note that LLMs are not yet adept at maths, and therefore caution should be exercised and large number-crunching is ill-advised.

Ensuring compliance with ever-changing tax legislation

Maintaining tax compliance can be an extremely complex task, particularly if a business operates in multiple countries and regions.

Additionally, the regular introduction of new legislation such as Pillar Two means that even the most diligent tax compliance processes can quickly become obsolete if they aren’t regularly reviewed and updated. The results of such errors can range from incorrect filings, paying the wrong amount of tax, or even incurring costly fines and reputational damage.

Here’s where AI models could change the equation. Not only could they be used to constantly monitor for changes in legislation that might otherwise go unnoticed, but they could also alert businesses when they’re about to cross tax thresholds for the first time or if changes to business operations suddenly bring new rules into scope.

When it comes to the power of AI technology, we’re only just beginning to scratch the surface of what’s possible. However, for industries like ours, where professionals grapple with time-consuming manual tasks on a near-daily basis, the potential benefits are already clear to see.

Replies (1)

Please login or register to join the discussion.

avatar
By FactChecker
02nd Jul 2024 00:56

"New AI-powered applications promise to make light work of such cumbersome processes. Not only can they effortlessly populate profit and loss (P&L) statements to an accuracy level of more than 90%, but by harnessing the power of generative AI, they can also prepare a tax treatment analysis to an accuracy level of more than 80%."

So cheaper than a trainee (maybe), but without any major discernible improvement in quality and accuracy ... and in the meantime where are the juniors (if still employed) going to get the experience and learning that will be needed in a few years' time?

Thanks (2)