Join the Movement

How AI Can Improve Brand Safety

AI is only as good as the data it gets, but it is as critical as the problem it's solving.

Most solutions for brand safety and suitability today are based on keyword filters: a text review tool that searches for specific words by exact match within a single message. 

But as brands are coming to realize, using a keyword filter to preserve brand safety is using a blunt instrument for a delicate operation. The problem, you see, is in interpreting context: for example, the word ‘shoot’ probably appears on a majority of keyword block lists, to prevent a brand from appearing adjacent to content about shooters, guns, or gun violence: seems sensible, right? 

However, many sports writers use the word ‘shoot’ to describe attempting a goal. A keyword list would block your brand from advertising near sports content - and you would miss out on reaching this large audience.

No matter how well-crafted, AI can only be as good - as accurate, reliable, and functional - as the data that is in the system. However, I would add that AI is also as critical as the problem it's solving. Using legacy tools to solve a complicated, evolving  problem like brand safety practically ensures that your response will be limited and ineffectual.

An AI model that is trained with integrity, using large amounts of excellent data can, on the contrary, increase the top-line of a business through brand safety and community retention. 

The differences between Contextual AI and Keyword filters as moderation tools are outlined below. 

While Contextual AI has obvious value over keyword filters as a solution, there are still challenges to creating and implementing an effective AI solution for brand safety.

The more ambiguity and corner cases that exist, the more data you need to correctly navigate those complexities. Like skiing -- if it were a straight shot down the hill, you’d only need one gate (Keyword match); if the path is twisting and curvy, you need more gates to direct the skier down the hill. Gates are the data we need here (by vertical, platform, use case and etc), and the path toward the destination -- your community guideline -- is the AI model, Contextual AI. 

While keyword filters are still commonly in use for managing brand safety, they are an imperfect solution. Advances in AI are promising as the next phase of brand safety, providing a contextual interpretation that can help brands move from brand safety - preventing damaging associations - to brand suitability - promoting positive associations.