SARS is checking more taxpayer information before tax returns are filed
Most South Africans only deal with the South African Revenue Service (SARS) when they submit a tax return. By then, the revenue authority has already received information from employers, banks, retirement funds and other reporting institutions. SARS now uses artificial intelligence, machine learning and data analytics to compare that information, identify discrepancies and decide which

SARS is checking more taxpayer information before tax returns are filed
Most South Africans only deal with the South African Revenue Service (SARS) when they submit a tax return. By then, the revenue authority has already received information from employers, banks, retirement funds and other reporting institutions. SARS now uses artificial intelligence, machine learning and data analytics to compare that information, identify discrepancies and decide which tax returns need verification or an audit. The technology forms part of the revenue authority’s effort to process growing volumes of taxpayer information without relying as heavily on manual reviews. “SARS uses data science, machine learning and AI as part of its broader modernisation programme to continuously innovate and improve tax compliance processes,” Siphithi Sibeko, Head of Communication and Media at SARS, told TechCabal. “The SARS strategy focuses on the customer experience and applying these capabilities to ensure that ‘tax just happens’.”
More compliance checks now start with automated risk assessments
SARS said its compliance programme generated R304 billion ($18.2 billion) during the 2024/25 financial year. Over the past five years, the revenue authority also stopped more than R417 billion ($25 billion) in invalid refund claims through AI-assisted fraud detection and verification. The systems also determine which cases officials investigate first. SARS said automated risk assessments now select every verification case and 88.41% of complex audit cases. That allows tax officials to spend more time investigating higher risk cases instead of reviewing large numbers of tax returns individually.
SARS compares tax returns with information from other organisations
Tax returns are only one source of information available to SARS. The revenue authority also receives records from employers, financial institutions, medical schemes, retirement funds, insurers, investment managers and other reporting entities. It combines those records with information from government departments, foreign tax authorities and cryptocurrency reporting frameworks where applicable. That allows SARS to compare the income taxpayers declare with information that other organisations submit throughout the year. On 1 July, SARS said it had auto assessed more than 1.9 million taxpayers and paid about R8 billion ($479 million) in refunds within 72 hours. The revenue authority said the latest updates are intended to make filing simpler and reduce the amount of information taxpayers need to submit themselves.
Online income gives SARS more information to compare
More South Africans now earn money through freelance work, online businesses, remote employment and cryptocurrency. Those activities usually generate records through payment platforms, banks, exchanges or other financial service providers. As more economic activity moves online, SARS has more third-party information available when reviewing tax returns. South Africa’s e-commerce market is expected to reach R130 billion ($7.8 billion), representing almost 10% of total retail sales. That growth also increases the amount of transaction data flowing through the tax system. “The focus is not on any single technology, but on a platform approach that integrates data, analytics, AI and modern compliance capabilities to make compliance easier for honest taxpayers and harder to evade for those who choose not to comply,” Sibeko said. He added that taxpayers must declare all taxable income, regardless of where or how they earn it.
Officials still review cases before SARS takes action
SARS says automated systems support compliance work but do not replace staff. According to Sibeko, officials review cases flagged by the system before the revenue authority takes any enforcement action. SARS also updates its models to improve accuracy and reduce false positives. The shift is less about replacing tax officials and more about helping them work through far larger volumes of taxpayer information. As more financial activity generates digital records, SARS can compare more information than would have been practical through manual reviews alone.



