Human culture has evolved for thousands of years and thrived in the era of Internet. Due to the availability of big data, we could do research on human culture by analyzing its representation such as user item rating values on websites like MovieLens and Douban. Industrial workers have applied recommender systems in big data to predict user behavior and promote web traffic. In this paper, we analyze the social impact of an algorithm named ZeroMat to show that human culture is locked into a state where individual's cultural taste is predictable at high precision without historic data. We also provide solutions to this problem and interpretation of current Chinese government's regulations and policies.