What's your contingency plan for someone using your service to process abusive/illegal content? It seems difficult to manage that scenario if it's a side project, and could easily get your API access banned.