AI fraud detection saved RM5.33 million in Malaysian employment incentive scheme
Malaysia's Social Security Organisation (PERKESO) used AI and data scrutiny to uncover a fraud scheme targeting the Daya Kerjaya 2.0 employment incentive programme, preventing approximately RM5.33 million from being misappropriated. The scheme involved employers claiming incentives for workers who were terminated immediately after the six-month benefit period ended or received only token payments despite full claims.
What happened
PERKESO's Anti-Fraud, Ethics and Integrity Division used artificial intelligence and data scrutiny to detect fraudulent claims in the Daya Kerjaya 2.0 employment incentive programme, beginning assessments and inspections at the end of 2024.
The investigation identified companies that appeared compliant but exploited the scheme by terminating workers immediately after the six-month incentive period ended. Fraudulent methods included submitting claims for workers who received only token payments, using personal information without consent, and misusing information of persons with disabilities for payments to other parties.
Information identified as having elements of fraud was handed to the Malaysian Anti-Corruption Commission (MACC) in August 2025 for enforcement action. As of June 17, 2026, RM1.26 million had been recovered from employers involved.
Reported impact
Through swift action by PERKESO's Anti-Fraud Division, approximately RM5.33 million was prevented from being siphoned off. Without these measures, losses could have reached about RM9 million.
Intelligence gathering and ongoing investigations identified 1,638 companies potentially linked to suspicious claims involving about RM45 million, with 143 companies believed to be involved in fraudulent claims of approximately RM9 million.
Index analysis
The AI system detected a pattern where employers submitted full incentive claims but terminated workers immediately after the six-month benefit period, suggesting systematic fraud rather than legitimate employment practices.
Governance controls
PERKESO indicated it would implement improvements recommended by MACC to close gaps that could be exploited. The agency stated that cooperation with MACC would continue, referencing previous joint operations against fraudulent PERKESO claims.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.