Malaysia National Fraud Portal — AI-powered real-time anti-scam network for financial intelligence
Malaysia
Bank Negara Malaysia (BNM) and PayNet's National Fraud Portal (April 2024) links 53 institutions via FNA graph analytics to trace …
Israel · Jerusalem · See the Israel profile
Israel's Tax Authority deployed AI to cross-reference capital declarations against financial data, identifying NIS 20 billion (~$7 billion) in discrepancies from a 4% filing sample. Plans to scale to 100% of all returns.
The Israel Tax Authority (Rashut haMisim) deployed AI-based tools that systematically cross-reference taxpayers' capital declarations against other financial data held in its systems, including bank records and investment data. Capital declarations — required of self-employed individuals and business owners — are compared to identify unexplained wealth gaps.
By June 2026, the system had identified discrepancies totalling NIS 20 billion (approximately $7 billion) from the roughly 4% of all filings it had scanned. At the Eli Hurvitz Conference on Economy and Society in Jerusalem (3 June 2026), Tax Authority Director Shay Aharonovich stated: 'Today we scan only 4%. We compared capital declarations in our systems with other reports and identified gaps totalling NIS 20 billion. We are moving toward reviewing 100% of all reports.'
The disclosure came as the Israeli government searched for ways to close a fiscal gap without raising taxes on the middle class. The 2026 revenue forecast was separately raised from NIS 540 billion to NIS 560 billion, partly in anticipation of additional enforcement yield from the AI-driven programme.
No technical specification of the AI system has been publicly released, and no independent audit or academic evaluation of the tool exists at the time of writing. The NIS 20 billion figure represents identified discrepancies and investigative leads rather than confirmed additional revenue collected. The system remains in active scale-up, with the authority committed to expanding from 4% to 100% coverage.
Read the full analysis: https://www.taxes.gov.il/English/Pages/default.aspx
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Where this practice's information was retrieved from, and when.
Malaysia
Bank Negara Malaysia (BNM) and PayNet's National Fraud Portal (April 2024) links 53 institutions via FNA graph analytics to trace …
Ireland
Since 2011, Ireland's Revenue Commissioners have used AI risk scoring (REAP) to target tax-compliance interventions; 290,000 data-analytics-driven interventions in 2023 …
United States of America
The US Treasury's Office of Payment Integrity uses ML to screen federal disbursements. In FY2024, ML-driven processes prevented or recovered …
Latvia
Latvia's SRS uses ESCORT ML to score every registered company for fraud risk across tax and labour domains. Fraud detection …
Open full copilot Grounded in cited practices — always check the sources.