Resemble AI gives Peacock a production-grade defense against the growing wave of deepfake audio targeting streaming platforms. Resemble Detect analyzes uploads and promotional content for AI-generated or cloned voices, catching fraudulent endorsements, misleading dubs, and deepfake clips before they damage brand trust.
PerTh neural watermarking embeds imperceptible ownership signals into every piece of audio Peacock publishes. The marks withstand re-encoding, pirate re-uploads, and third-party AI training making it possible to trace unauthorized distribution back to its source and defend licensed IP at scale.
Run Detect across uploads and user-submitted content to identify AI-generated voices impersonating NBC talent or characters.
Embed neural watermarks into every show, film, and trailer. Prove ownership and detect piracy with frame-accurate precision.
Watermarks survive compression, re-encoding, screen recording, and AI model training — near 100% recovery rate.
Shield on-air personalities and licensed characters from voice-cloning fraud and unauthorized AI redubs.
Scan audio as it enters ingestion. Fits into existing QC and publishing workflows without slowing time-to-air.
SOC 2 Type II (in progress), GDPR, and on-prem options. Audio stays within Peacock's compliance boundary during processing.
