Fake news has become a major problem in the cyberspace with far reaching consequences. The open access and unregulated social networks are popular attack vectors that are frequently used to propagate misinformation. To fight this problem, several naïve solutions have been proposed, including a blockchain implementation of news feed to distinguish facts from fiction. However, the size of social networks and the design constructs of blockchain add several new challenges that impede the real world deployment of such solutions. In this paper, we postulate a new blockchain system that overcomes the existing challenges and limits the spread of fake news across the network. Towards that, we analyze the information workflow in the social networks and construct an optimal detection system that can be effectively deployed with minimal overhead. Moreover, our proposed solution can be extended beyond social networks to other online platforms.