Redakcja Choose TV
The rise of generative AI has made creating realistic videos easier than ever. At the same time, it’s becoming harder to tell real footage from AI-generated material. The problem of deepfakes is an increasing challenge for media outlets, companies, and internet users. NVIDIA has responded to this issue by unveiling a new AI tool for detecting manipulated videos. The solution was showcased at the SIGGRAPH 2026 conference. NVIDIA emphasizes that the technology is meant to support the verification process, not replace human work.
Synthetic Video Detector analyzes videos in real time
The new NVIDIA tool is called Synthetic Video Detector and was developed to detect material generated or altered by artificial intelligence. The system analyzes footage frame by frame and then assigns a probability that AI was used in creating the video. According to NVIDIA, analyzing Full HD footage takes just 22 milliseconds on computers equipped with RTX GPUs. This allows the solution to be used practically in real time. The tool has been made available as part of the NVIDIA NIM Microservices platform, so organizations can easily integrate it into existing systems. The company stresses that the technology is meant to serve as an additional layer of protection for newsrooms, broadcasters, and enterprises. Its purpose is to quickly verify material before publication. NVIDIA emphasizes that it does not replace traditional journalistic analysis or the methods used by digital forensics specialists.
High effectiveness should help fight the growing wave of disinformation
According to data presented by NVIDIA, the new solution reaches effectiveness of up to 92 percent when analyzing uncompressed video. For footage compressed by about 15 percent, accuracy falls to 87 percent, while with stronger compression of 50 percent it reaches around 82 percent. Compression remains one of the biggest challenges for deepfake detection systems, because platforms such as YouTube, TikTok and Instagram automatically reduce the quality of uploaded content. NVIDIA also says that Synthetic Video Detector achieved top results in the AI GVD Bench benchmark, used to compare tools that detect synthetic video. The company also plans to integrate the technology with the Wowza Intelligence Video Framework, which operates in more than 35,000 deployments across 170 countries. The manufacturer stresses, however, that the tool is not a perfect solution and should support, not replace, human evaluation of material. As generative artificial intelligence advances, the sophistication of fake footage is also increasing. Combating disinformation will therefore require the simultaneous development of detection technologies and responsible human work.
The Synthetic Video Detector demonstrates that artificial intelligence can be used not only to create realistic content but also to detect it. With very fast performance and high accuracy, NVIDIA's new solution could become an important tool in the fight against deepfakes and online disinformation. The company stresses, however, that even the best algorithms cannot replace careful human verification of information.
source: digitaltrends
