1Kosmos & Reality Defender unite to combat deepfakes

Identity verification firm 1Kosmos has partnered with deepfake detection specialist Reality Defender to add real-time analysis of synthetic media into digital onboarding and authentication checks.

The integration embeds Reality Defender's technology inside the 1Kosmos platform. Customers can run deepfake detection during live video sessions, on recorded images and video, and alongside existing document and biometric checks. Both companies position the move as a response to rising fraud that uses AI-generated faces, voices and videos in remote identity processes.

Industry data from TransUnion shows suspected deepfake-driven fraud rose nearly 700% year-on-year. The report cites financial services, telecommunications, government services and online marketplaces as key targets. It also states that 20% of biometric fraud attempts now involve deepfakes.

Deepfake focus

The combined service extends the presentation attack detection that 1Kosmos already uses for liveness and spoof checks. Presentation attack detection standards sit under ISO/IEC 30107-3, which vendors use as a benchmark for biometric fraud resistance.

Reality Defender's tools analyse different media types. These include video, audio and still images. The aim is to flag synthetic or manipulated content during identity verification without changing existing user workflows.

"Advances in AI-generated impersonations are rewriting the rules of identity assurance and ratcheting up fraud losses," said Mike Engle, Chief Strategy Officer for 1Kosmos. "By adding Reality Defender as an embedded detection layer, we're enabling enterprises to verify identity with greater certainty and stop AI-driven impersonation attacks before they result in financial loss, brand damage, or regulatory consequences."

The partnership covers both live and pre-recorded submissions during onboarding and authentication sessions. It sits inside the 1Kosmos platform, which the company says processes millions of authentications per day for more than 75 million users worldwide.

Regulatory pressure

Vendors in identity verification and fraud prevention face growing regulatory and compliance expectations. Policymakers in several regions are drafting or updating rules that address AI risk, synthetic media and biometric privacy.

The companies say the joint service aligns with existing and emerging standards. These include ISO/IEC 30107-3 and the planned ISO 25456 standard. They also reference the EU AI Act, which sets obligations around high-risk AI systems and transparency requirements for synthetic media.

1Kosmos states that customers can add the new deepfake checks without separate licences or major technical changes. The integration works within existing identity workflows and user interfaces. The firms say that more accurate early detection can reduce the volume of manual reviews and fraud investigations and can lessen compliance workload.

"Deepfake attacks are evolving faster than most organizations can adapt, and detecting them requires specialized, continuously updated models," said Ben Colman, CEO of Reality Defender. "By integrating our technology directly into the 1Kosmos platform, customers gain uncompromised protection against synthetic media within their existing workflows, achieving stronger identity assurance at scale."

Customer uptake

The combined service is available now. 1Kosmos says large retail and telecommunications customers already use the integrated deepfake detection in production environments.

Both companies market to sectors that run remote onboarding or high-volume account access. These include banking, telecommunications, technology providers, healthcare organisations and retailers.

Reality Defender supplies its detection engines through a web platform and an API. The firm uses a multi-model approach that targets different types of AI-generated media, including video, audio, imagery and text.

1Kosmos provides remote identity verification and passwordless multi-factor authentication. It links identity proofing, credential checks and strong authentication in a single platform.

The firms say they will continue to update detection models as new types of deepfake attack emerge and as regulatory rules develop in different regions.

"By integrating our technology directly into the 1Kosmos platform, customers gain uncompromised protection against synthetic media within their existing workflows, achieving stronger identity assurance at scale." said Colman.