The Montgomery Summit• Mar 16, 2026• 22:24InterviewInterview

Authenticity Infrastructure for the AI Era: Why Organizations Need Scalable Deepfake Detection

From The Montgomery Summit · 2026

Milos Fulton•Chief Operating Officer, Reality Defender

Executive Summary

The threat from AI-generated deepfakes has rapidly evolved from a theoretical curiosity to a severe, active threat for enterprises, with real-world attacks bypassing identity verification, targeting contact centers, and enabling hiring fraud.
Reality Defender provides an API-first "authenticity infrastructure" to detect deepfakes, integrating directly into enterprise workflows like contact centers (JPMorgan), video conferencing (Zoom/Teams), and critical infrastructure security (Port of Los Angeles via IBM).
The company maintains its technological edge through a research-first approach, with a majority of its staff being PhDs and engineers, developing proprietary foundation models, and leveraging unique data partnerships (e.g., ElevenLabs).
A new, emerging threat is the use of agentic AI to conduct large-scale, Denial-of-Service (DDoS) style attacks against contact centers, indicating a new vector of AI-driven cyber threats beyond simple impersonation.

9 quotes

Concerns Raised

The ease and speed (under 15 minutes) with which convincing deepfakes can now be created.
The imminent risk of a massive, publicly-disclosed breach involving deepfakes, which will act as a tipping point for the industry.
The use of agentic AI to conduct large-scale DDoS attacks against critical infrastructure like contact centers and 911 services.
Nation-state actors (e.g., North Korea) are actively using deepfakes to commit hiring fraud and fund their regimes.

Opportunities Identified

Becoming the essential 'authenticity infrastructure' layer for enterprises by integrating into existing high-trust workflows.
Securing major enterprise verticals, especially financial services, government, and healthcare, which are high-value targets.
Leveraging strategic partnerships with major players like IBM and JPMorgan to accelerate enterprise adoption and co-develop solutions.
Expanding the developer ecosystem and discovering novel use cases through a free, public-facing API.

Key Themes

The Escalating Threat of AI-Generated Fraud

The discussion highlights the increasing severity and accessibility of deepfake technology, demonstrated by the ease of creating a convincing deepfake in just 15 minutes. This has led to a rise in sophisticated attacks, including bypassing identity verification, executive impersonation for financial fraud, and nation-state actors committing hiring fraud.

This signifies a critical shift where deepfake attacks are no longer a future problem but a present-day reality, requiring immediate attention and defensive solutions from all enterprises, particularly in finance and government.

Enterprise Adoption of Deepfake Detection

Reality Defender is seeing significant traction in regulated industries, with major deployments in financial services (JPMorgan), government, and healthcare. Key use cases include securing contact centers from impersonation, adding a deepfake detection layer to identity verification (IDV) systems, and preventing hiring fraud in remote work environments.

The adoption by major institutions like JPMorgan validates deepfake detection as a necessary component of the modern enterprise security stack, moving it from a niche tool to a core infrastructure requirement.

Building 'Authenticity Infrastructure' via APIs

The company's strategy is to provide an API-first platform that acts as an underlying "authenticity infrastructure." This allows for easy integration into existing enterprise systems like Microsoft Teams, Zoom, and contact center software, rather than forcing clients to adopt new, standalone workflows.

This API-centric approach is crucial for scalability and market penetration, as it allows for rapid deployment across diverse, pre-existing enterprise environments and encourages a developer ecosystem to build new solutions on top of the platform.

The AI Security Arms Race

To stay ahead of rapidly evolving generative AI models, Reality Defender operates as a research-first organization, building its own proprietary detection models. The company employs a multi-pronged data strategy that includes generating its own data, partnering with generative AI companies like ElevenLabs, and purchasing high-quality datasets.

This highlights the continuous, research-intensive investment required to combat AI-driven threats, underscoring that off-the-shelf solutions are insufficient and a deep, foundational R&D capability is a key competitive differentiator.

Emergence of Agentic AI as a Threat Vector

A forward-looking threat discussed is the use of agentic AI to bombard and overwhelm enterprise systems. Specifically, attackers can use AI agents and auto-dialers to effectively launch a Denial-of-Service (DDoS) attack on a contact center, crippling its ability to serve legitimate customers or respond to emergencies.

This points to the next evolution of AI-powered attacks, moving beyond one-to-one impersonation to large-scale, automated system disruption, which presents a new and significant national security and infrastructure risk.

Get started free

Topics

Deepfake Detection AI Security Generative AI Enterprise AI Identity Verification (IDV)Contact Center Security Hiring Fraud API-first Agentic AI Cybersecurity Financial Services Government Technology National Security On-premise Deployment Venture Capital

Processed Apr 19, 2026 yt-dlp + mlx-whisper + Gemini