IAS announces industry-first deepfake measurement capabilities
Integral Ad Science, a global media measurement and optimization platform, has announced availability of the industry’s first deepfake measurement offering, enabling advertisers to further verify the safety and suitability of their digital media investments through trusted and safe AI-powered technology.
Now in Beta testing, the new offering helps advertisers avoid running adjacent to deepfake content as part of the Global Alliance for Responsible Media (GARM)-defined Brand Safety Floor and Suitability Framework misinformation category.
Lisa Utzschneider, CEO, IAS: “Marketers are looking to leverage new technologies to detect and mitigate the harmful effects of deepfakes and misinformation, and we’re excited to offer this latest innovation ahead of the U.S. elections. Marketers need reliable partners and data to help them safeguard their brands, and our deepfake detection offering will empower them with the actionable data needed to address this rising threat.”
Deepfake signals, combined with IAS’s misinformation categorization capability as part of IAS’s industry-leading brand safety products, use AI-driven Multimedia Technology with frame-by-frame analysis to provide the most reliable and accurate data in the industry. IAS’s latest innovation aims to detect visual manipulations, including faceswaps in videos and images synthesized by Generative AI technologies, and to protect the authenticity of content aligned to the GARM framework.
Kumaresh Singh, SVP, Data Science, IAS: “IAS Multimedia Technology safeguards and scales brands across channels, with up to 74% greater accuracy. We are dedicated to benchmarking our deepfake detection technology against cutting-edge open source frameworks and models along with rigorous testing within our own classification framework. We are excited to continue developing video-first solutions in close partnership with our customers.”
