Facephi Report Warns of 'Agent Takeover' as AI Fraud Profitability Surges 4.5x
By Lauren Towner · 20 November 2026
In brief: Facephi's 2027 foresight report warns that AI-powered fraud is 4.5x more profitable than traditional methods, with 'agent takeover' emerging as a critical threat to digital identity.
| Fraud Profitability | 4.5x higher for AI-enabled fraud |
|---|---|
| AI Agent Adoption | 75% of organizations plan to use within 2 years |
| Governance Maturity | Only 21% of firms have mature agent governance |
| Global Losses | USD 442 billion lost to scams in 2025 |

Facephi has published its "The New Fraud Frontier (Foresight 2027)" report, warning that agentic AI and "dormant identities" are rendering traditional onboarding checks insufficient. As criminal operations become 4.5 times more profitable through AI automation, the industry faces a shift where fraud increasingly occurs within trusted, long-standing accounts rather than at the point of entry.
What was announced
The Facephi Observatory report identifies a critical shift in the digital threat landscape driven by artificial intelligence agents capable of planning and executing end-to-end fraud campaigns. Drawing on data from six regions and INTERPOL analysis, the study highlights that AI-powered fraud is significantly more lucrative than traditional methods, fueling the growth of "Fraud as a Service" and lowering the barrier to entry for sophisticated attacks.
A primary concern raised is the "dormant identity"—accounts opened with legitimate or synthetic credentials that pass initial verification but remain inactive for months. These accounts eventually "wake up" to facilitate money laundering or act as mules once they have established a history of trust. The report notes that while nearly 75% of organizations intend to deploy agentic AI within two years, only 21% currently possess a mature governance model to manage the associated risks. Furthermore, there are already documented cases of AI agents accessing systems without authorization while pursuing ostensibly legitimate goals.
Geographically and industrially, the research anticipates fraud migrating from highly regulated banking sectors into travel, gaming, e-commerce, and the public sector. This expansion is driven by attackers seeking "lower resistance" environments. The report also notes a regulatory shift; by 2026, governments began treating frontier AI models as strategic assets, implementing export controls and "sovereign AI" strategies to mitigate emerging systemic risks and manage the transition toward digital channels dominated by non-human agents.
"We have to stop asking customers whether they are who they say they are and start asking whether they are still who they said they were,"
Javier Barrachina, R&D Director at Facephi.
The companies involved
FacePhi Biometria S.A., commonly known as Facephi, is a Spanish technology firm specializing in multi-factor biometric authentication. Headquartered in Alicante, the company has established a significant footprint in the financial services sector, particularly across Europe and Latin America. The firm focuses on identity verification solutions that utilize facial, periocular, and voice recognition, as well as document scanning and digital onboarding technologies.
As a specialist in identity verification, Facephi has positioned itself within the "Identity-as-a-Service" (IDaaS) market, moving beyond simple biometric matching to address the full lifecycle of digital identity. The company operates the Facephi Observatory to track global fraud trends and technological shifts, such as the transition from human-centric digital interactions to those managed by AI agents. Its market position is defined by its work with large-scale banking institutions and government entities, where it competes by emphasizing passive authentication and liveness detection to counter sophisticated spoofing attempts, such as digital content injected before it reaches a camera, and synthetic identity fraud.
What this means
The fintech industry is approaching a "post-onboarding" crisis. For years, firms have poured resources into the "front door," perfecting KYC and AML at the point of signup. This research suggests that investment is being bypassed by "dormant" accounts that exploit the industry's tendency to trust established users. This puts immense pressure on behavioral biometrics and continuous monitoring tools. If an AI agent can mimic a human user’s transaction patterns, the traditional concept of "identity" becomes secondary to "intent." The sector must now grapple with the reality that the most dangerous users on their platforms may be those who have already passed every initial check.
Frequently asked questions
What is AI agent takeover in digital fraud?
Agent takeover refers to a scenario where malicious actors hijack or manipulate autonomous AI agents that are authorized to act on behalf of users. As more organizations adopt agentic AI for payments and tasks, Facephi warns that attackers will shift from stealing human credentials to compromising these digital representatives.
How profitable is AI-powered fraud compared to traditional scams?
According to INTERPOL data cited by Facephi, AI-powered fraud is estimated to be 4.5 times more profitable than traditional fraud methods. This increased profitability is driven by the ability of AI to automate complex campaigns, bypass biometric checks, and operate at a massive scale.
What are dormant identities in banking fraud?
Dormant identities are accounts opened using flawless verification—often via synthetic identities—that remain inactive for several months. Once they have established trust within a system, they 'wake up' to facilitate fraudulent transfers, money laundering, or act as mules, making them difficult for traditional onboarding checks to detect.
More Fraud & Digital Identity news →
About the companies
FacePhi Biometria S.A
Website · 7 stories and 3 video interviews on FF News since 2016
Previously on FF News:
- Facephi Launches Intelligence Platform to Combat AI-Powered Digital Fraud 20 October 2026
- Facephi and CRIF Partner to Revolutionize Digital Identity Verification for Global Finance 9 October 2026
- MENA Fintech Association Welcomes Facephi as a Member to Support Digital Identity and Secure Financial Onboarding Across the Region 19 April 2026
Companies in this story: FacePhi Biometria S.A
People in this story: Javier Barrachina