{
  "id": 11197277,
  "title": "AI-powered tax scams target South Africans ahead of SARS deadline",
  "url": "https://urgent.news/2026/10/01/ai-powered-tax-scams-target-south-africans-ahead-of-sars-deadline",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-01T13:15:00.000Z",
  "source": {
    "name": "IOL",
    "slug": "iol",
    "url": "https://iol.co.za/the-star/news/2026-10-01-cloned-ai-powered-tax-scams-target-south-africans-ahead-of-sars-deadline/"
  },
  "original_language": "en",
  "account": "As the October 23 tax filing deadline looms, South Africans are increasingly falling victim to AI-powered scams targeting their personal and financial data. Cybercriminals are leveraging the growing public trust in Artificial Intelligence, crafting increasingly convincing fake messages, investment offers, and tax documents that mimic the South African Revenue Service (SARS).\n\nExperts warn that AI technology has made it exceedingly difficult to distinguish between genuine and fraudulent communications, with high-quality deepfakes and sophisticated phishing attempts now commonplace. Anna Collard, SVP Content Strategy & CISO advisor at KnowBe4 Africa, and Chris Norton, general manager for Sub-Saharan Africa at Kaspersky, express concern that scammers are exploiting the urgency surrounding tax refunds, assessments, and outstanding matters to pressure victims into acting swiftly without verifying the legitimacy of the message.\n\nSARS has issued warnings about scams and phishing attempts during the 2026 tax season. Research indicates that 62% of South Africans are willing to trust AI, compared to a global average of 46%. Additionally, 86% of South African respondents believe deepfakes are so realistic that it's challenging to discern what to trust.\n\nCollard notes that scammers no longer rely on obvious red flags such as poor grammar, fake logos, or unnatural voices. She explains that \"people accept that the threat is real and hard to detect, and they just don't believe it applies to them.\" She emphasizes that trust should be based on the channel rather than the content, stating that \"trust should attach to the channel, not the content. A video of someone you recognise, a message in a familiar voice, a polished PDF, none of it is evidence of anything anymore.\"\n\nCollard also cautions against relying on AI chatbots for critical financial decisions, stating that \"AI tools are fluent, and we mistake fluency for expertise. Use these tools to understand concepts and formulate better questions. Don't use them as your adviser.\" Norton highlights the approaching tax deadline as a prime opportunity for scammers, pointing out that phishing websites mimicking SARS eFiling and malicious attachments with names like \"Tax Returns Of RXX,XXX.XXX\" and \"Final_Assessment_Notification-2026-0X-0X.pdf\" have been identified by Kaspersky researchers. These fake emails often use phrases like \"Refund due\" and \"SARS Letter of demand\" to entice recipients into opening attachments or following instructions.\n\nSARS explicitly warns taxpayers that they will never request sensitive information such as passwords, one-time PINs, or eFiling login credentials through email, SMS, social media, or telephone. Norton advises taxpayers to be cautious about providing information to AI services regarding their tax matters, as AI tools can process detailed personal and financial data. He warns that sensitive records should remain within trusted sources and stresses the importance of verifying crucial tax information through official channels rather than clicking on links in unsolicited messages.",
  "summary": "Experts warn that AI is making fake SARS messages, refund notices and tax documents harder to detect as South Africa’s October 23 filing deadline approaches.",
  "key_points": [
    "AI scams target South Africans ahead of October 23 tax filing deadline",
    "Cybercriminals use AI to create convincing fake messages and documents mimicking SARS",
    "86% of South Africans believe deepfakes are realistic, making it hard to discern trust"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}