Two leading deepfake detection companies announced a partnership that merges audio and visual fraud detection into a single platform, helping organizations catch sophisticated scams that span phone calls, videos, and fabricated documents. Scam.ai, which specializes in detecting manipulated images and videos, is integrating Modulate's synthetic voice detection technology to create what they call the first unified multimodal deepfake defense system.

Why Are Scammers Using Multiple Channels at Once?

The fraud landscape has fundamentally changed. According to research from Gallup and the Stop Scams Alliance, an estimated 15.1 million U.S. adults were personally scammed in 2025, resulting in at least $68 billion in losses. The same study found that phone calls, text messages, and email were each involved in 45% of scams, with roughly half of all scams crossing two or more communication methods.

This shift matters because traditional detection tools were built to examine one type of content at a time. A fraudster might start with a cloned voice call to establish urgency and trust, then follow up with a fabricated image or document to reinforce the deception. If your fraud team is only monitoring videos, they miss the voice component. If they're only listening to calls, they miss the visual manipulation.

"Scammers stopped limiting themselves to one channel a long time ago, but many detection systems are still organized around individual media formats," said Dr. Ben Ren, co-founder and CEO of Scam.ai.

Dr. Ben Ren, Co-founder and CEO of Scam.ai

What Does the New Integrated Platform Actually Do?

The partnership brings Modulate's synthetic voice detection directly into Scam.ai's existing platform. Customers will be able to analyze live or prerecorded audio alongside images and videos, receiving confidence scores and detection signals that indicate whether a voice is synthetic or AI-generated. The integrated capability is expected to be available through Scam.ai in early September.

Modulate's voice detection model reports 98.9% accuracy and a 1.1% equal error rate, and holds first place on the Hugging Face Speech Deepfake Detection Leaderboard as of August 4, 2026. Scam.ai's visual detection models report 98.2% accuracy against the company's internal benchmark. Together, they create a system that can catch deepfakes across the three primary forms of synthetic media in a single workflow.

How to Implement Multimodal Deepfake Detection in Your Organization

• Unified Platform Approach: Deploy a single integrated system that analyzes audio, video, and images together rather than maintaining separate detection tools for each media type, reducing complexity and improving detection accuracy across coordinated fraud attempts.
• Confidence Scoring Integration: Use confidence scores and detailed detection signals to prioritize high-risk content for additional human review, allowing your team to focus resources on the most suspicious interactions.
• Workflow Embedding: Add synthetic voice detection to existing fraud prevention, authentication, and content-verification workflows without requiring staff to learn new systems or change established processes.

The partnership addresses a critical gap in how organizations currently defend against fraud. As Carter Huffman, CTO and co-founder of Modulate, explained, the problem is fundamentally about timing and coordination.

"People are being asked to determine whether a voice, image or video is authentic at the exact moment a scammer is trying to manipulate them. That is an adversarial problem, and detection cannot depend on whether someone thinks a voice sounds suspicious," said Huffman.

Carter Huffman, CTO and co-founder of Modulate

Where Will This Technology Be Used First?

The integrated platform has broad applications across industries. Potential use cases include identity verification and digital onboarding, financial fraud and payment authorization, executive and employee impersonation, contact center security, social media and user-generated content moderation, insurance claims, digital evidence verification, and enterprise cybersecurity investigations.

For financial institutions, the timing is critical. A fraudster might call a customer with a cloned voice claiming to be from the bank, then send a fabricated document requesting wire transfer authorization. With multimodal detection, security teams can flag both the suspicious voice and the manipulated document as part of the same incident, rather than treating them as separate alerts.

The partnership reflects a broader industry recognition that deepfake attacks no longer fit neatly into single-format categories. As scams become more sophisticated and coordinated across channels, the detection systems designed to stop them must evolve to match that complexity. The Scam.ai and Modulate integration represents one of the first commercial attempts to solve this problem at scale, offering organizations a practical way to defend against multimodal fraud without deploying multiple disconnected tools.