By CPA Simon Ndirangu
Artificial Intelligence (AI) has the potential to considerably improve the way the Kenya Revenue Authority (KRA) monitors taxpayers’ activities and spots any suspicious behaviour, such as tax evasion. One of the great things about AI is its ability to process and analyze vast amounts of data in real time. Using machine learning algorithms, the KRA can examine financial data from various sources, including bank transactions, mobile payments like M-Pesa, and business records. This analysis helps them spot patterns and identify inconsistencies in tax reporting. For instance, AI can emphasize these discrepancies for further investigation if a taxpayer’s declared income doesn’t align with their transaction history or a business’s reported sales are unusually low compared to industry standards. With this level of real-time monitoring, it becomes much more complicated for taxpayers and companies to hide income, inflate expenses, or under-report their revenue.
AI can considerably boost the KRA’s ability to identify taxpayers who might be dodging their taxes. By sifting through historical tax data and looking at how taxpayers typically behave, AI can help determine those individuals and businesses at a higher risk of evasion. This more innovative, more focused strategy allows KRA to direct its audit efforts toward the most likely problematic cases instead of conducting broad, costly audits. By honing in on these high-risk areas, this approach helps save money on enforcement and increases the likelihood of revealing any non-compliance. Plus, with machine learning, these algorithms can adapt over time, continually improving their accuracy as they process new information.
AI’s ability to monitor transactions in real time gives the KRA a powerful tool for keeping an eye on financial activities as they unfold. With AI at their disposal, they can easily spot large or unusual transactions that might signal tax evasion. For instance, if a business suddenly shows a big spike in revenue without a matching rise in expenses or filings, AI can flag this for the KRA to investigate further. This kind of immediate oversight not only helps catch potential tax evasion quickly but also sends a strong message to taxpayers: the KRA is vigilant and attentive. This proactive approach can be a strong deterrent against evasion.
A major portion of Kenya’s workforce operates in the informal sector, largely outside the reach of traditional tax collection methods. Many of these workers and businesses avoid paying taxes simply by not registering with the Kenya Revenue Authority (KRA) or by failing to declare their earnings. This is where AI can make a difference. By analysing electronic transaction data from widely used mobile money platforms like M-Pesa, which is incredibly popular in Kenya, AI can help the KRA identify informal businesses. By looking at factors such as transaction volumes, patterns, and business activities, AI can determine these businesses and encourage them to join the tax system. By bringing more businesses into the fold, we can expand the tax base, allowing for a fairer contribution to national revenue.
Kenya, like many other nations, is struggling with dodgy practices such as shell companies, revenue under-reporting, and trade mispricing. This is where AI comes in. It has the potential to spot these evasive tactics by analysing business formations, transaction histories, and ownership details. With the help of machine learning algorithms, we can discover unusual patterns that might suggest using shell companies or fraudulent financial statements, which can then be investigated further. Besides, AI can track cross-border tax evasion, revealing how some businesses take advantage of international trade and offshore accounts to dodge their tax responsibilities. By examining global trade patterns and identifying inconsistencies in imports and exports, AI can help determine businesses that are downplaying their transactions or mispricing goods to sidestep taxes.
AI Tools to Enhance Taxation in Kenya
To accurately use the power of AI in the tax system, the Kenya Revenue Authority (KRA) has a wealth of tools and platforms at its fingertips, each capable of optimising various aspects of the tax process. For instance, OpenAI’s ChatGPT is a handy AI-powered chatbot ready to assist taxpayers with their queries. This technology can handle common questions about taxes, guide individuals and businesses through the filing process, and provide real-time updates on their tax obligations. By integrating ChatGPT into KRA’s official platforms, the Authority can offer round-the-clock support, easing the workload on human staff while enhancing the overall experience for taxpayers. Also, ChatGPT can deliver educational materials that break down complex tax laws, helping individuals understand how to stay compliant with ease. Another powerful option is IBM Watson, which offers advanced capabilities for data analysis and predictive modeling. KRA could use Watson’s skills to develop machine learning models that identify potential non-compliance based on historical data and transaction patterns. Thanks to Watson’s cognitive computing abilities, the KRA could better assess risk, prioritise cases for audits, and make knowledgeable enforcement decisions grounded in solid information. Microsoft’s Azure AI platform is another excellent resource. With its cloud-based data analysis and machine learning capabilities, KRA could analyse vast amounts of information from diverse sources, including banking transactions, mobile payments, and tax reports. By incorporating Azure into its existing systems, the KRA could establish a stronger data environment where AI algorithms spot anomalies and help reveal potential tax evasion. What’s more, Google Cloud’s AI functionalities, like AutoML and TensorFlow, enable KRA to create personalized machine learning models that can swiftly determine suspicious financial activities. By integrating these AI tools with mobile payment systems such as M-Pesa and various banking transaction systems, the KRA can rapidly detect signs of tax evasion. Google Cloud provides a scalable infrastructure that ensures the capability to handle large datasets, allowing the KRA to monitor transactions in real time across the nation. Lastly, SAP Leonardo offers a smart solution for automating business processes. This AI and machine learning tool could simplify the KRA’s operations by automating routine tasks like processing tax returns, verifying documents, and matching invoices. What’s more, it can identify irregularities in financial reports that might indicate tax evasion or underreported income. By implementing SAP Leonardo, the KRA can enhance its efficiency, minimize manual errors, and accelerate data processing. Incorporating these advanced tools will considerably support the KRA in achieving its goals, making the tax system more efficient and fairer for everyone involved.
The Path Forward: Steps for Successful AI Implementation
The potential for AI to revolutionise Kenya’s tax system is considerable, but turning that promise into reality demands careful planning and investment. To effectively integrate AI into its tax processes, the KRA needs to take several important steps.
1. Investment in Technology and Infrastructure
We need to prioritise our investment in technology infrastructure. This involves acquiring data storage systems, using AI tools, and enhancing our capacity to manage large volumes of transactional data. Also, the KRA needs to focus on developing secure and efficient digital platforms that can effectively collect and analyse data in real-time.
2. Building Capacity and Training Personnel
For AI to be effectively integrated into taxation, it’s essential to have the right people on board to manage and utilise these technologies. The Kenya Revenue Authority (KRA) will need to focus on developing its internal capabilities, either by hiring new talent or by training existing staff in areas like data science, machine learning, and AI-driven analytics. This approach will enable the KRA not only to operate AI systems but also to continually refine them, helping to identify emerging trends and innovative methods of tax evasion.
3. Public Education and Transparency
For AI-driven tax systems to gain widespread acceptance, we need to engage the public and clarify how AI operates and its importance in promoting fair taxation. The Kenya Revenue Authority should launch campaigns aimed at educating people about the advantages of AI, particularly in eliminating tax evasion and guaranteeing that everyone plays their part in the nation’s growth.
Conclusion
AI presents a fantastic opportunity for the Kenya Revenue Authority (KRA) to considerably enhance its tax collection processes, curb tax evasion, and improve overall efficiency. By using AI’s strengths in data analysis, predictive modeling, real-time monitoring, and risk-based auditing, Kenya can encourage a taxation system that’s not only more effective but also fairer for everyone involved. Tools like ChatGPT, IBM Watson, Google Cloud AI, and Microsoft Azure AI are essential assets that can simplify the country’s tax collection efforts. With thoughtful planning, strategic investment, and careful implementation, AI has the potential to truly change the KRA’s tax framework.
The writer is accounting and data analytics professional currently working as an Accountant/Bookkeeper at Avicenna Accounting Automation, Inc. in Virginia, USA. With expertise in financial reporting, taxation, auditing, and data-driven decision-making, he leverages analytical tools like SQL, R, Tableau, and Power BI to optimize financial and business strategies. Additionally, He utilizes AI systems to automate accounting processes.
symondirangu29@gmail.com
