Silverpush, a leading AI-powered technology solutions company, today announced the launch of their new brand suitability platform Mirrors Safe. This product forms a part of the company’s computer vision-enabled video intelligence product suite, which is transforming how some of the largest brands globally reach their most engaged audience in a brand-safe environment.
Brands today face a significant financial risk from a potential crisis involving their advertising. A 2019 study by Trustworthy Accountability Group (TAG) and Brand Safety Institute (BSI) found that 80% of customers will stop or reduce buying products that place advertisements across harmful or offensive content, and around 70% blame the brand or the agency for such placement. This survey exhibits the real and measurable risk that a preventable brand safety crisis can pose.
Through the use of AI-powered in-video context detection, Mirrors Safe transcends challenges of traditional brand safety measures like keyword-based technologies and whitelisted channels. Trained with millions of pieces of visual content, the platform identifies in-video contexts like objects, actions, faces, and scenes within a streaming video, and contextually filters out harmful content across an extensive set of brand unsafe categories including nudity, smoking, violence, crashes, and much more.
Mirrors Safe’s advanced algorithm uses five parameters to calculate a comprehensive brand suitability score, that measures not only safety and suitability of content, but also of the page and the channel. These include:
- Engagement: likes, dislikes & participation that the content generates.
- Safety: exclusion through in-video context detection, on-screen text, and audio sentiment analysis.
- Influence: organic influence that channel/page/content creates.
- Relevance: how relevant is the content in terms of its peer channel/page category.
- Momentum: consistency that channel/page maintains or grows in terms of engagement.
Kartik Mehta, CRO at Silverpush said, “What sets Mirrors Safe apart is its ability to custom define the scope of harmful contexts that are unique to every brand. Thus, helping brands move beyond just brand safety to a truly brand-suitable environment. This is limited with existing keyword and natural language processing (NLP) based blanket exclusion technologies, as these often fail to understand the complex undertones and various contexts words can be used for”.
In a testing phase, Silverpush analysed ~15 million videos across the largest video hosting and sharing platforms in South East Asia using Mirrors Safe. Overall, 8-9% of content was found to be brand unsafe across multiple categories, indicating that nearly 1 in every 9 video ad placements are placed around harmful and damaging content. Top unsafe content categories identified include smoking, violence, adult, and extremist content.
To further showcase the impact context-relevant brand safety can bring in, the company used both traditional measures and Mirrors Safe’s in-video context detection to identify content for one of the top brand unsafe categories – nudity and adult content. Mirrors Safe, through its frame by frame parsing of video content, identified 300% more videos featuring in-video unsafe contexts in this category, compared to traditional methods.
Even a single misplaced ad can harm brand perception for a long time in the consumer’s mind– this discovery points out the inefficacy and potential harm that existing measures can have, which has been witnessed time and again from continued failures of some of the largest video advertising platforms in keeping brands safe.
Kartik further added, “Mirrors Safe further addresses one of the most pressing brand safety challenges of content over-blocking – a result of blanket exclusion measures offered today. This significantly limits campaign performance and often forces marketers to switch off controls in favour of reach. Mirrors Safe’s in-video context detection technology prevents over-blocking and only filters videos that actually feature unsafe contexts, ensuring brand safety without hampering monetisation and performance.”