Overview of the use of AI for tax administration

Let me create a blog post on the use of Artificial Intelligence (AI). With reports of the use of ChatGPT in politics and in the courtroom, the use of AI in tax administration has become a common trend as technology advances at an accelerated pace in today's world.

In this post, we will discuss our current efforts to leverage AI to improve our tax operations. So without further ado, here you go!

Disappearing tax attorney stories due to technological developments

Advances in technology have transformed the work of tax professionals. In the past, professionals did everything from manually entering paper tax returns to calculating taxes, reporting, and preparing applications. Today, however, tax professionals need to leverage technology in order to focus on more sophisticated tasks.

The story that "tax accountants are the first profession to disappear" conversely provides tax accountants with a lot of fodder. Like the argument that won't disappear, or asking an AI a tax question. Well, that is delicious, so it may not be a bad idea to have tax accountants labeled as they are.

The "Housing Debate! Renters? Buyer?" This story is likely to continue on a regular basis as well.

What does AI mean here?

AI here means artificial intelligence. It has a wide range. An overview of what is popular these days shows that "weighted answers using neural networks" plus "natural language answers" are the impact.

In the first place, a large percentage of non-specialists are involved in taxation, so it is understandable that the use of AI in tax administration is expected to improve the efficiency of operations.

However, since human judgment is required regarding applicable rules and tax law interpretations for individual cases, the AI's main role is to support human judgment. even if you ask ChatGPT, the last thing it says is "Ask an expert. We believe this is the result of learning considerations that do not harm a particular group of experts.

I can see how issues such as the impact on employment and ethics associated with the use of AI will come up at the same time as efficiency increases. I would like to follow this area with great interest.

Using AI to Improve Efficiency in Tax Administration

Let's look at some examples of how AI is being used to improve the efficiency of tax administration.

AI to identify undeclared and delinquent taxpayers

In tax administration, the use of AI is making it possible to identify taxpayers who have failed to file returns or are delinquent. Tax collectors are introducing AI to identify taxpayers with a high possibility of undeclared tax returns and to approach delinquent taxpayers with the aim of improving taxpayer convenience and increasing the efficiency and sophistication of taxation and collection. Specifically, by extracting, processing, and analyzing various information inside and outside the national tax organization, and analyzing taxpayers' personalities and tendencies, taxpayers who fail to file returns or are delinquent can be identified and dealt with at an early stage. In addition, the system is expected to be economically effective, as it can significantly reduce the time and effort required compared to conventional tax administration.

Simplifying tax returns and the role of AI

The increasing use of AI technology in tax returns is expected to increase convenience for taxpayers. Specifically, this includes e-Tax use and automatic checking of tax return contents. It is a bit of a question as to whether this technology is AI. It could be said to be mere computing. Nevertheless, the use of AI is expected to make it possible to identify taxpayers who have failed to file tax returns or who are in arrears, thereby increasing tax revenues.

Legal issues of tax administration due to AI replacing expert judgment also need to be discussed. This is an area of interest to see if this could be a topic for a tax law paper.

AI and data analysis for advanced survey and collection

We believe that the use of AI technology will lead to more sophisticated investigation and collection operations in tax administration.

One of our long-established colleagues said, "The Electronic Bookkeeping Act was intended to be introduced for large companies, but it was forced to go to small and medium-sized companies. I am happy if the situation is settled in such a way.

This is because it is conceivable that AI-based analysis could allow data to be read and processed all at once. In the Electronic Bookkeeping Act, the requirement to save all data seems to be a milestone. Although there are opinions within the agency and the movement will be slow, the technology will make it much easier to do this kind of thing.

AI and natural language processing to handle tax inquiries

The IRS is already working on this.Report on Fiscal 2020It is also listed on the Usage will continue to grow, and the use of AI and natural language processing to respond to tax inquiries is one of the most advanced technological innovations in tax administration. Overseas, the use ofPwC leverages AI's natural language processing technology to develop chatbot capabilities to quickly respond to internal inquiries.Its use is already underway, for example.

This will significantly improve operational efficiency by allowing AI to handle the inquiry response tasks that were previously handled by numerous tax staff members. In addition, the use of AI is expected to contribute to improving the credibility of tax administration, as AI, with its solid judgment as a tax expert, will be able to provide high-quality answers and explanations by utilizing its vocabulary, logical analysis, and other skills.

Japan has not confirmed this, but in the U.S. we try to keep tax-related data limited to servers in the U.S.

Tax AI's impact on estate taxation and valuation formulas for unlisted companies

The impact of tax AI on the estate tax system and the valuation method of unlisted companies is expected to be significant in the U.S. AI has already been introduced in the U.S. Estate tax is a tax that is paid by heirs based on their inherited assets, and while it is sometimes difficult to assess the tax, it is expected that AI's data analysis and inference technology will enable more accurate and fair assessments.

The same applies to the valuation of unlisted companies. It is technically possible, given that Google engineers say that inferences can be made without being as big as big data.

I am not sure if the figures will be convincing. It may be that after the number of cases of introduction overseas increases inexplicably, discussions about the introduction of the system in Japan may emerge.

Legal Issues in Tax Administration Brought on by AI

The introduction of AI into tax administration is expected to improve efficiency and accuracy, but at the same time, legal issues are emerging. For example, algorithms related to false declarations and management of confidential information may violate personal information protection and tax laws.

To comply with legal regulations, AI technology must be combined with legal knowledge to ensure proper monitoring and management. But can the project proceed with that level of holistic understanding? There are going to be many legal issues. If the choice to stop walking away from the project rather than risk problems occurs, the use of AI will be less likely.

summary

The use of tax AI is advancing rapidly. You may be surprised at the applications that you would never imagine crossing the threshold. This is an interesting field and I would like to follow it.