In a bid to protect brands from associating their ads with objectionable content, digital network Fullscreen has unveiled today a proprietary tool called Fullscreen Shield.
The cross-platform software leverages image and audio recognition to analyze a video’s metadata, tags, titles, comments, and audience ratings — a vastly more sophisticated approach given that most screening tools typically focus on thumbnail imagery, according to the company. These findings are then audited against Fullscreen’s database of unsafe terms. (Fullscreen says Shield harnesses machine learning to constantly discover new unsafe words as they enter the cultural zeitgesit).
Brands can choose to spend according to content that has been rated by Fullscreen per the above calculations. Prior to the launch of a campaign, Shield can also trigger a human review for videos that are potentially unsafe.
Subscribe for daily Tubefilter Top Stories
In addition to its brand safety features, Fullscreen Shield can also identify campaign-specific content so that ads are better targeted in terms of relevancy and recency, Fullscreen says. All told, the tool maps content against more than 25 categories and 12,000 total keywords. The tech works on YouTube videos (across fields like animation and gaming), and Facebook and Instgram videos as well.
“Five billion videos are watched each day, and the industry lacks the tools needed for a simple but comprehensive way to automatically screen content,” Mark Williams, Fullscreen’s senior director of media operations, said in a statement. “Brand safety is not universal — there is no single list that incorporates all unsafe content as each list is brand-specific, which is why Fullscreen Shield combines machine learning and human review.”
This isn’t the first time that Fullscreen has taken pains to ensure advertisers of brand safetly. In April, the company released approximately 160 creators from their partnership contracts, citing concerns about inappropriate content.
