Infrastructure Platforms

AI-enabled fraud forces overhaul of identity verification systems

Wednesday, 10 December 2025 7:40AM UTC

As AI-driven scams proliferate with convincing synthetic identities and deepfakes, industry leaders warn that legacy verification methods are collapsing, prompting a shift towards continuous, enterprise-wide fraud prevention strategies to safeguard global commerce and public safety.

Verification systems built for yesterday’s fraud are collapsing under a new onslaught of AI-enabled attacks, forcing firms to treat identity and fraud prevention as enterprise-wide, continuously evolving problems rather than siloed, annual reviews. According to the original report and interviews conducted by PYMNTS with industry figures, cheap generative models now produce convincing synthetic identities, deepfakes and automated fraud swarms that routinely defeat human inspection and legacy checks. ^[1]^[2]

Research cited in the PYMNTS piece quantifies the scale: identity gaps now drain more than 3% of global revenue , roughly $95 billion a year , even as many organisations remain overconfident in their ability to spot automated threats. The report found 96% of firms expressed confidence in detecting harmful bots despite nearly 60% struggling to do so in practice, a mismatch that executives from Trulioo and WEX said underscores the accelerating pace of attack innovation. ^[1]

Industry leaders interviewed described a seismic shift in operational tempo. “The sophisticated tooling to defraud a system … is now available to a much wider swath of bad actors,” Zac Cohen, chief product officer at Trulioo, told PYMNTS, while William Fitzgerald, vice president of Global Fraud & Financial Crimes at WEX, warned: “The barrier to entry into becoming a fraudster at scale is essentially gone.” Those comments reflect a broader consensus that monthly, if not continuous, reassessments of identity controls are now required. ^[1]

The technical manifestations of the threat are visible across sectors. Third-party data and platform studies report dramatic rises in AI-enabled fraud: a payments and identity vendor warned of synthetic identity document fraud surging by more than 300% in the United States, while a separate identity firm reported deepfake attacks occurring with alarming frequency and digital document forgeries growing steeply year-on-year. Survey data from verification providers similarly show one in three organisations has already been hit by identity spoofing or biometric fraud. These figures reinforce the PYMNTS conclusion that the problem is widespread and growing. ^[5]^[6]^[7]

The risk extends beyond commerce to public safety and trust. A recent local policing incident in Kansas demonstrated how voice deepfakes can trigger emergency responses: a caller using an AI-cloned voice convinced a relative that a kidnapping had occurred, prompting a police mobilisation before the deception was uncovered. Experts quoted in coverage of the case said law enforcement and legal frameworks are struggling to keep pace with such synthetic-media scams. That episode exemplifies the real-world consequences regulators and firms fear if verification lags behind attack capabilities. ^[3]

International bodies have echoed the urgency. A United Nations-affiliated report presented by the International Telecommunication Union called for stronger measures to detect AI-driven deepfakes, urging companies and platforms to adopt advanced detection tools and digital verification systems to curb misinformation, election interference and fraud. The ITU statement reinforces the argument that responses require both private-sector technology upgrades and coordinated regulatory action. ^[4]

Executives argue the economic case for modern identity is not just loss avoidance but growth enablement. Better identity flows, they say, improve user experience, reduce false positives and accelerate customer approval , outcomes that lift conversion and lifetime value. Conversely, manual review bottlenecks and intrusive checks can depress expansion, particularly across borders where know‑your‑business (KYB) systems are ill-suited to non‑domestic registries and paperless corporate forms. The PYMNTS interviews and accompanying analysis frame fraud prevention as a strategic enabler, not merely a cost centre. ^[1]

Practically, experts recommend a layered, adaptive architecture: central governance with local configurability; parallel deployment that complements rather than immediately replaces legacy systems; and the construction of “trust graphs” that synthesise registry records, web presence, behavioural signals and other indicators to spot anomalies invisible to single-point checks. According to the original report, these trust graphs and continuous verification approaches often deliver measurable lift , Cohen cited typical market improvements in the 20–30% range , helping to justify incremental migration away from brittle legacy controls. ^[1]^[2]

The confluence of vendor data, investigative reports and international warnings paints a clear picture: the tools to commit fraud at scale are now widely available, deepfakes and synthetic identities are already inflicting material harm, and the response must be coordinated, continuous and technically sophisticated. Industry data and regulatory appeals both point toward the same conclusion , identity must be engineered for an adversary that can now generate convincing fakes at scale. ^[5]^[6]^[4]^[7]^[3]

📌 Reference Map:

##Reference Map:

^[1] (PYMNTS) - Paragraph 1, Paragraph 2, Paragraph 3, Paragraph 7, Paragraph 8, Paragraph 9
^[2] (PYMNTS summary) - Paragraph 1, Paragraph 8
^[3] (Axios - Lawrence, Kansas) - Paragraph 5, Paragraph 9
^[4] (Reuters/ITU) - Paragraph 6, Paragraph 9
^[5] (PR Newswire - Sumsub) - Paragraph 4, Paragraph 9
^[6] (Entrust) - Paragraph 4, Paragraph 9
^[7] (GlobeNewswire/Regula) - Paragraph 4, Paragraph 9

Source: Noah Wire Services

More on this

https://www.pymnts.com/fraud-prevention/2025/ai-fraudsters-crash-identity-systems-built-for-yesterday/ - Please view link - unable to able to access data
https://www.pymnts.com/fraud-prevention/2025/ai-fraudsters-crash-identity-systems-built-for-yesterday/ - This article discusses how AI has lowered the barrier for large-scale fraud by making synthetic identities, deepfakes, and automated fraud swarms more accessible and harder to detect. It highlights the need for enterprises to shift to continuous and adaptive defenses powered by modern verification systems to manage emerging global identity gaps. The piece also emphasizes the importance of integrating diverse signals into identity frameworks to create 'trust graphs' capable of mapping and staying ahead of evolving threats.
https://www.axios.com/local/kansas-city/2025/12/04/ai-voice-scam-lawrence-police - This article reports on a recent incident in Lawrence, Kansas, where a woman was convinced her mother had been kidnapped based on a convincing AI-generated voice call. The call, which mimicked her mother's voice using audio likely pulled from social media or voicemail greetings, prompted a police response that ultimately revealed the threat was fake. Experts say that AI-generated fraud—including deepfakes and synthetic identities—is outpacing police training and legal frameworks.
https://www.reuters.com/business/un-report-urges-stronger-measures-detect-ai-driven-deepfakes-2025-07-11/ - This article discusses a UN report released by the International Telecommunication Union (ITU) at the 'AI for Good Summit' in Geneva on July 11, 2025. The report warns of the increasing risks posed by AI-generated deepfakes and calls for urgent action. The ITU emphasizes the need for companies and social media platforms to adopt advanced detection tools and digital verification systems to combat misinformation, election interference, and financial fraud driven by deepfakes.
https://www.prnewswire.com/news-releases/synthetic-identity-document-fraud-surges-300-in-the-us--sumsub-warns-e-commerce-healthtech-and-fintech-at-risk-302479872.html - This press release from Sumsub reveals a dramatic rise in AI-enabled fraud across the United States. According to platform data, deepfake fraud has surged by 1100%, while synthetic identity document fraud rose by over 300%, with attackers exploiting generative AI to create fake passports, IDs, and biometric data. The release highlights the need for businesses to adopt advanced identity verification solutions to protect against these sophisticated fraud attempts.
https://www.entrust.com/company/newsroom/deepfake-attacks-strike-every-five-minutes - This article discusses findings from Entrust's 2025 Identity Fraud Report, revealing that deepfake attacks occurred every five minutes in 2024, while digital document forgeries increased 244% year-over-year. The report emphasizes the importance of digital identity verification in preventing fraud and financial crime, noting that this first moment of interaction provides organizations with the opportunity to establish trust in an individual's identity from day one.
https://www.globenewswire.com/news-release/2025/09/17/3151595/0/en/Deepfakes-Already-Hitting-Businesses-as-Often-as-Traditional-Fraud-Regula-Survey-Finds.html - This article reports on a survey by Regula, a global developer of identity verification solutions, showing that identity spoofing, biometric fraud, and AI-powered deepfakes have already struck one in three organizations worldwide, catching up with long-standing fraud tactics like forged documents and social engineering. The survey highlights the need for businesses to invest in advanced identity verification solutions to protect both their operations and customers.

Noah Fact Check Pro

The draft above was created using the information available at the time the story first emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed below. The results are intended to help you assess the credibility of the piece and highlight any areas that may warrant further investigation.

Freshness check

Score: 8

Notes: The narrative was published on December 5, 2025, making it current. The earliest known publication date of similar content is May 13, 2025, with Plaid updating its identity verification product to combat generative AI-powered fraud. ([pymnts.com](https://www.pymnts.com/fraud-prevention/2025/plaid-updates-identity-verification-product-to-combat-generative-ai-powered-fraud?utm_source=openai)) The report is based on recent interviews and research, indicating high freshness. No evidence of recycled content or discrepancies in figures, dates, or quotes was found. The narrative includes updated data, justifying a higher freshness score. ([pymnts.com](https://www.pymnts.com/fraud-prevention/2025/ai-fraudsters-crash-identity-systems-built-for-yesterday/?utm_source=openai))

Quotes check

Score: 9

Notes: The direct quotes from Zac Cohen and William Fitzgerald are unique to this report. No identical quotes appear in earlier material, indicating original content. The wording matches the original sources, with no variations found. No online matches were found for these quotes, suggesting exclusivity. ([pymnts.com](https://www.pymnts.com/fraud-prevention/2025/ai-fraudsters-crash-identity-systems-built-for-yesterday/?utm_source=openai))

Source reliability

Score: 8

Notes: The narrative originates from PYMNTS, a reputable organisation known for its coverage of financial and payment industry news. The report cites interviews with industry leaders and references recent research, enhancing its credibility. No unverifiable entities or fabricated information were identified. ([pymnts.com](https://www.pymnts.com/fraud-prevention/2025/ai-fraudsters-crash-identity-systems-built-for-yesterday/?utm_source=openai))

Plausibility check

Score: 9

Notes: The claims about AI-driven fraud and its impact on identity systems are consistent with recent industry reports. For instance, Sumsub's annual report highlights a shift towards more sophisticated fraud schemes in 2025. ([prnewswire.com](https://www.prnewswire.com/news-releases/sumsubs-annual-report-fraud-shifts-to-complex-multi-step-schemes-in-2025-agentic-ai-scams-poised-to-surge-in-2026-302625287.html?utm_source=openai)) The narrative includes specific data points and quotes from verifiable sources, supporting its plausibility. The language and tone are appropriate for the topic and region. No excessive or off-topic details were found, and the tone is consistent with typical corporate language. ([pymnts.com](https://www.pymnts.com/fraud-prevention/2025/ai-fraudsters-crash-identity-systems-built-for-yesterday/?utm_source=openai))

Overall assessment

Verdict (FAIL, OPEN, PASS): PASS

Confidence (LOW, MEDIUM, HIGH): HIGH

Summary: The narrative is current, original, and originates from a reputable source. The claims are plausible and supported by verifiable data. No significant issues were identified, indicating a high level of credibility.

AI fraud
identity verification
cybersecurity
deepfakes
regulation