 There's still a lot of hype around AI and it's hard to separate the hype from the reality. It's still the wild wild west of AI and it's an exciting time but people need to do work and not look for an easy simple solution. We really have really invested all of our energy into computer vision specific part of AI which is basically teaching computers to see. We've combined that with our natural language processing capabilities of AI which is text and for us we've created a practical application around brand safety. There's so much value in artificial intelligence today and the problems that we are able to solve and so I think that's probably one of the biggest marketing challenges we face is separating that hype from the reality. We've always believed that the future of advertising is going to be within the content other than outside the content and the advantage to our industry when it comes to AI is that we now have more structured data than any other marketing industry. What it is in fact is actually the new electricity right. AI is going to power everything that we do and I think that's really exciting but it's not again you're not going to apply it to these big sweeping changes. What you what you really want to do is think about what is the discrete problem I'm trying to solve. We need to try to prevent ourselves from falling into that little trap of wait for something to happen and react and I think we need to take a more aggressive stance in our partners and especially the advertisers and brands that can really own the dollars going into the ecosystem to really push for adoption and quicker adoption of innovative technologies to really help make sure that if they're going to spend the money it's being spent not only where they want it to be but first and foremost in the safe place that it should have been in the very beginning. We have to use the data the right way and in a very secure way and that's why we've decided to develop our technology internally. We have our own data scientists we have our own deep learning experts where we create these models and these algorithms to make sure it's very controlled and that our clients are not compromised. This shouldn't be for AI for AI's sake this is about what is the fundamental business problem we need to solve.