Dalet today announced it is working with NVIDIA to bring advanced synthetic content detection directly into media workflows. The new capabilities integrate NVIDIA AI models with Dalet Flex, enabling media organizations to identify potentially AI-generated images and video as content enters their operations and incorporate those insights into existing review, approval and governance processes.

Dalet said the work builds on its approach to bringing AI closer to the media itself, embedding intelligence directly into the infrastructure and workflows organizations already use to manage their content. By combining NVIDIA AI models with Dalet Flex and Dalet AI services, organizations can analyze content where their media resides, including on-premises and hybrid environments, while connecting detection insights directly to the operational workflows where decisions are made.

“As synthetic content becomes more sophisticated and more prevalent, knowing what is entering your media operation is becoming fundamental to maintaining trust,” said Erwan Kerfourn, Head of Product, Dalet. “That need is taking on even greater importance as requirements such as the EU AI Act introduce new transparency obligations around AI-generated and manipulated content. Working with NVIDIA allows us to bring powerful detection capabilities directly into the media workflow, giving teams the information and human oversight they need to make informed decisions about the content they use and publish.”

The EU AI Act, which became broadly applicable on August 2, 2026, introduces transparency requirements for certain AI-generated and manipulated content. Article 50 50 requires providers of AI systems that generate synthetic audio, image, video or text to make those outputs machine-readable and detectable as artificially generated or manipulated, and establishes disclosure requirements for deployers in certain circumstances, including the use of deepfakes.

Dalet brings NVIDIA Synthetic Video Detector directly into Dalet Flex, turning AI analysis into actionable intelligence within the media workflow. NVIDIA models analyze images, individual video frames, and video content for signs of AI generation or manipulation by diffusion models, with the resulting intelligence surfaced within Dalet Flex.

The synthetic content detection capability in Dalet Flex is currently available to Dalet customers for proof-of-concept (POC) deployments.

“For media operations such as news, the challenge isn’t simply detecting synthetic content; it’s knowing what to do with that information inside a real production workflow,” adds Erwan. “Our early work with news customers shows the value of bringing NVIDIA’s detection capabilities directly into Dalet Flex, where potentially AI-generated or manipulated content can be flagged for further review and human validation. We’re using that real-world feedback to refine the offering and, together with NVIDIA, explore additional AI models that can address the specific needs of media organizations.”

Dalet will present its workflows and technologies at IBC 2026.

“As AI-generated content becomes more prevalent, media organizations need trusted tools to understand and manage what enters their workflows,” said Richard Kerris, vice president of media and entertainment, NVIDIA. “By integrating the NVIDIA Synthetic Video Detector AI model into Dalet Flex, Dalet brings synthetic content detection directly into media operations, helping teams identify AI-generated material and make informed decisions about how it is reviewed and used.”