Can India take the big step by combining artificial intelligence with taxation? The major source of revenue for the Government is through taxes. So, can we move towards a system wherein AI will help taxation system to be clear, easy and more compliant? Extensive use of AI & ML will definitely help our tax authorities to be vigilant about the defaulters and create a more comprehensive and transparent system. Can India move towards a taxation system where AI/ML is used extensively in the functioning of tax systems in the country? As per recent sources, India is the first country to go for faceless tax assessment system.
To manage the overall changing scenario in taxes, the country is looking towards the increased use of data analytics by exploring different avenues in AI so as to facilitate compliance and assist professionals in various activities like filing of returns, solving queries, analysing the behaviour & psyche of wilful defaulters etc. Artificial Intelligence, one of the best technologies of Industry 4.0 is transforming every area how countries and businesses operate. AI is helping companies and Governments to manage risk in a better manner.
Technology can process volumes of information in a fraction of a second. AI & tax together have immense potential if channelled in a correct manner. This study aims to understand the implications of AI/ML in the area of taxation and how AI can help in creating a better risk bearing and compliant system
The study is carried out on the basis of secondary data. Data was collected through secondary sources. The research method used was literature review and library research was used for analysis (Secondary data from websites and research papers)
Benjamin Franklin had said that “nothing can be certain in this world except death & taxes” but considering so many developments taking place in the field of technology and AI , we can definitely add AI to this list.
Artificial Intelligence is concerned with making smart machines capable of performing tasks or work which require human intelligence. AI is not only about reducing the manpower or eliminating human beings, but it is a way in which the role of humans is increased/augmented in the work they perform.
This paper helps in understanding the role of AI in the field of taxation. Taxation involves large amount of data – structured and unstructured data coming from various sources. AI & big data is one of the ways through which the inflow of information is channelized.
Artificial Intelligence Concept
AI uses automation to replicate human intelligence to improve the analysis and decision making capabilities. These systems can sense, comprehend, act. AI and application utilize automation to enhance or replicate human intelligence to improve the decision-making capabilities of machines. It helps in completing complicated tasks easily and effectively.
Development of AI in taxation
Taxation functions can be improvised with software robotics and there can be a digital transformation. It can reduce the work and enhance the workload of over 350 tax agents in 1 year. Income Tax Departments are planning to use Big Data to catch tax evaders. Some of the problems with India is failure to pay tax dues, submission of false tax returns, lack of reporting income, income earning & storing in foreign countries.
Sometimes, many people purchase products or indulge in lavish lifestyle beyond monetary means as per their tax returns. This is mostly not visible to the tax officials and they are not able to assess and analyse everything on a manual level. But it is easy to analyse this information with Big Data and Data analytics technologies. Tax laws in India was created in 1961, but there is a rapid need for change in today’s times. There is a need for effective tax management.
Example of AI in taxation – Till now, the income tax officer used to inquire by asking for documents. But now if the system will find any income with no declaration then it will lead for deep inquiry of the person. For example, suppose someone has a TDS deduction of 2 lakh and that is not declared in ITR then the AI and ML based system will proceed for inner scrutiny of the person. It will scrutinize a person based on the e-verification and through system designed questionnaire. Consequently, the government will integrate all information regarding tax from CBDT, MCA, and GS after the declaration of this information to the public.
India is one the countries which uses Big Data to track tax evasion. A country with huge population like India cannot only rely on traditional methods of survey. Many of the analytical problems could not be solved before due to high expenses. Also, the analytical platforms could not handle such large volumes of data. The data collected were not present in the electronic form. All these problems made it difficult to manage and analyse large amounts of data. Therefore, we need Big Data to analyse an astronomical amount of data that is generating every day.
It is a combination of Big Data Management System and Machine Learning. It can help understand the income and expenditure patterns and inconsistencies in these two. This will be analysed with the IT returns and addresses of the tax payers or assessees. This can help in reducing tax evasion to some extent.
With launch of National e-Assessment Centre (NeAC), the Income Tax Department will usher in a paradigm shift in its working by introducing faceless e-assessment to impart greater efficiency, transparency and accountability in the assessment process. AI can help in basic as well advanced ways. The basic form filling or Form 16 which is got from the employer needs to be deduced and the same figures have to appear in the ITR while filing the returns. But when a person or assessee in unable to understand which amount has to be put where in ITR, it can be a tough task. Hence basic functions of AI can automatically link the Form 16 with ITR, which makes it easy for the assessee to understand and at least file the returns.
The second advanced function which data analytics and AI can handle is to understand the defaulters mindset or psyche and take care of critical data which needs to be entered. The 26 AS needs a bit of overhaul and it has to be simplified to again incorporate the TDS items in Form 16 to make it easier for assessee to fill it. The assessee should be easily able to understand and incorporate standard deduction from salaries
Many tax professionals perform the same data again and again. But Robotic processing can help avoid the same and automate the process and prefill numbers wherever required. It can help in reconciliations, consolidations and updating the financial data.
Machine Learning can help in extracting key facts and data from computer system. It can guide an assessee as to use which ITR, it can prefill data from bank accounts and TDS amounts. It can help in showing the deductions to be claimed and how much more deductions can be availed. The Income Tax liability can also be calculated. AI can use predictive modeling that can identify potential fraud. It can help in finding clues that are hidden in huge data and not able to be tracked by auditors.
How AI is developed in India
For India to maximally benefit from AI revolution, it should adopt policies to drive AI innovation, adaptation in different sectors which affect the economy in a large way. The education sector needs to be revamped to a better suitable AI powered economy for the future. Public Sector authorities and applications like detecting tax fraud, subsidy leakages and defence research needs to be restructured where AI can make a significant impact.
AI technology has a huge potential to shape India’s economic and national security. The policy makers need to tap this area where they can design policies for short and long term and thereby realize the potential for new technologically and technically advanced India. The only thing that they need to be vigilant about is the detrimental effects of AI proliferation.
As it is said, AI is the new UI; deep learning and machine learning – it can be used for as intelligent process automation, enhanced judgement, enhanced interactions. The technologies can be put to use in terms of language – content classification, natural language search, machine translation, emotion detection, language detection, in terms of vision and sound- object recognition, face recognition, object tracking, optical character recognition, emotion detection, scene recognition, speech to text, speech recognition, emotion detection, language detection etc.
Tax authorities around the world are facing many challenges brought along with international requirements for full transparency, substance and consistency of data worldwide. Leveraging AI as a tool to enhance efficiency, useful outliers and data analytics for the tax administrations has also been a trend among countries.
The US tax code becomes more complex and difficult to understand even for highly educated professionals. In the United States, IRS has been using Big Data on its phone surveillance records, tracking social media accounts and using extensive data mining to develop analytical algorithms for identification of tax compliance issues.
- Australia has partnered with Accenture to create and implement integrated information technology platform – Integrated Core Processing (ICP) and Siebel. ICP is used for client
- China has made AI driven technology the primary motive of its economic development.
- The UK Government had implemented big data analytics in their income tax segment and observed an increase of $ 5.4 billion in tax return revenue.
Some Problems with AI implementation in India
If there are job losses, it will surely impact the revenues and in turn will affect the countries’ tax revenues. If Government wants to encourage technological innovations, it has to find ways where the revenue will not reduce but rather increase in terms of real income It is a difficult task to balance the two things, if robots empower humans to create value it will be surely be an artificially intelligent world but with no human touch.
As sophisticated technology presents new opportunities and changes how things are done, training and hiring employees with high-tech skills becomes a necessity. One of the important aspects of AI is data sharing, but how well is the confidentiality of the information protected? Solution to this challenge is one of the key aspects of AI in taxation.
AI system needs adjustments in order to produce accurate tax audit report. Keeping the system up to date with new policies will make the AI system more acceptable. However, at present, AI is still developing, and the AI system is not able to update itself since manual adjustments are needed.
AI has lot of advantages if used wisely by the policy makers to balance the world economy. AI definitely can be used to make tax forecasting and tax compliance more accurate. AI applications need to be trained by people who understand and know tax. Integrating AI into the tax function can help improve efficiency and help in meeting the growing digital demands of tax administrations.
- Huang, Z.W. (2018) Discussion on the Development of Artificial Intelligence in Taxation. American Journal of Industrial and Business Management , 8, 1817-1824. https://doi.org/10.4236/ajibm.2018.88123
Websites & Pdf files
- Anon, (2019). [ebook] Available at: http://the-data-intelligent-tax-administration-whitepaper.pdf
- Volkenburgh, Brandon Van. “Artificial Intelligence and Taxes: 8 Ways It’s Being Used.” www.crowdreason.com/blog/artificial-intelligence-tax.
- India & Artificial Intelligence Revolution.pdf
- Andrej Kovacevic-Andrej Kovacevic- Andrej- Andrej- Fintech – https://readwrite.com/2019/03/22/ai-is-changing-how-we-pay-taxes-and-how-theyre-collected-too/
- How Ai Looks To Upend the Tax Function’s 80/20 Rule EY Global – https://www.ey.com/en_lu/tax/how-ai-looks-to-upend-the-tax-function-s-80-20-rule
- Using Ai To Power Prosperity During Tax Season The Hive – https://medium.com/hivedata/using-ai-to-power-prosperity-during-tax-season
- Accenture & Artificial Intelligence.pdf
- PWC (2015) Unlocking the Power of Data and Analytics. https://www.pwc.com/gx/en/tax/publications/assets/PwC TFoF Data Analytics global nov2015.pdf.
- Milner, C. and Berg, B. Tax Analytics—Artificial Intelligence and Machine Learning—Level 5. https://www.pwc.com/us/en/services/tax/tax-innovation/artificial-intelligence-andmachine-learning.html