A peculiar cruelty now accompanies sophisticated artificial intelligence scams targeting Americans' retirement accounts. Victims lose their life savings to convincing fraudsters, only to discover weeks later that the Internal Revenue Service considers those stolen funds taxable income. This double blow—financial devastation followed by an unexpected tax obligation—reveals a critical gap between modern fraud realities and century-old tax regulations.
Courney Werning, principal attorney at Meyer Wilson Werning, has witnessed this pattern escalate dramatically among her clients. Perpetrators deploy AI-generated deepfakes, cloned voices, and professional-grade investment platforms to manipulate victims into withdrawing funds from Individual Retirement Accounts and 401(k)s under false pretences. The withdrawals are treated as legitimate distributions by the tax authorities, regardless of whether the funds were stolen. "The financial loss is devastating," Werning explains, "but many victims don't realise the tax consequences can make the damage even worse."
The scale of this problem has exploded alongside AI capabilities. The Federal Trade Commission documented reported fraud losses reaching US$12.5 billion in 2024, more than five times the US$2.4 billion recorded in 2020. The FBI's Internet Crime Complaint Center separately reported Americans lost over US$16.6 billion to cybercrime in 2024, with investment fraud generating the largest category of losses. These figures represent not mere statistics but thousands of families facing retirement insecurity and psychological trauma.
What makes this situation particularly punitive is how retirement withdrawals are taxed under existing law. When funds are extracted from IRAs or 401(k)s, the Internal Revenue Service classifies them as taxable income distributions, regardless of the circumstances. Victims find themselves owing taxes on money that criminals successfully stole from them, creating an absurd outcome where the government effectively benefits from the fraud. For those under retirement age, an additional 10 percent early withdrawal penalty compounds the injury, sometimes forcing victims to owe more to authorities than the actual amount lost.
The modern sophistication of AI-enabled scams defies the email-based fraud victims may historically have encountered. Werning describes cases where scammers invest months building relationships and trust, often combining romantic manipulation with investment opportunities. In one instance, a victim was sent what appeared to be a legitimate US$100,000 check, reinforcing confidence in the scheme before the perpetrators requested increasingly substantial withdrawals from retirement savings. The psychological manipulation is deliberate and calculated, exploiting human psychology rather than merely deploying technical deception. These aren't obvious scams; they're meticulously constructed illusions that convince intelligent, cautious people.
Beyond financial metrics lies the psychological devastation these frauds inflict. Werning notes that many victims experience overwhelming shame, depression, and isolation after realising they've surrendered decades of retirement savings to criminals. Some experience such severe emotional consequences that the psychological burden becomes unbearable. The loss represents not merely money but security, independence, and dignity—elements that define retirement quality for millions of Americans approaching or already in their later years.
Recognising this injustice, bipartisan legislators have introduced the Tax Relief for Fraud Victims Act (HR 9500) to address what legal experts consider an unintended consequence of outdated tax code. If enacted, the legislation would restore theft-loss deductions for qualifying fraud victims, waive the 10 percent early withdrawal penalties when retirement funds are stolen, and permit taxpayers to amend returns retroactively based on when the fraud actually occurred rather than when it was discovered. These reforms acknowledge that tax law written for different eras fails to account for modern realities where sophisticated criminals can convincingly impersonate investment professionals and financial institutions.
For Malaysian and Southeast Asian observers, this situation carries relevant implications. The region has experienced similar surges in AI-enabled fraud, with perpetrators exploiting cross-border capabilities to target victims across multiple jurisdictions. Malaysian authorities have documented increasing reports of investment scams involving deepfakes and cloned voices targeting citizens. The American tax consequence scenario may not directly apply under Malaysian tax law, but the underlying vulnerability—how retirement savings are uniquely susceptible to sophisticated fraud—mirrors regional patterns. If similar frauds increase in Malaysia, regulators must consider whether existing frameworks adequately protect retirement account holders and whether tax treatment of stolen funds fairly reflects modern fraud realities.
Werning emphasises that documentation becomes critical for victims hoping to claim future protections should the legislation pass. Preserving bank records, wire transfers, communications with scammers, and law enforcement reports creates the evidentiary foundation necessary for legal remedies. For victims already experiencing financial devastation and emotional trauma, this administrative burden adds another layer of difficulty, yet represents an essential step toward eventual recovery or tax relief.
The fundamental problem emerging is that tax regulations written for simpler economic eras fail to address how modern AI-enabled fraud operates. When criminals successfully steal retirement funds through convincing deception, the current system punishes victims twice—once through the theft and again through taxation. This perverse outcome has galvanised bipartisan support for reform, reflecting recognition that current law produces manifestly unjust results. Lawmakers are essentially racing to modernise regulations faster than scammers can weaponise emerging technologies.
As artificial intelligence capabilities expand and become more accessible to bad actors, the gap between fraud sophistication and legal protections will only widen without deliberate legislative action. The Tax Relief for Fraud Victims Act represents a necessary first step, but ongoing vigilance will be required to ensure that future technologies don't create new loopholes exploiting elderly Americans and retirement savers. The cruel irony of current law—that victims owe taxes on money criminals stole from them—demands urgent correction before more Americans experience this compounding catastrophe.
