Not sure - there are a few databases of indeterminate origin floating around the internet, but as we're a) located in the litigation-friendly US and b) large enough to possibly be worth someone's trouble to sue, we have to be scrupulous about this stuff, so we can really only use an open-source database with a clear copyright trail like CC-CEDICT. (this is also why we've been unable to use the extremely popular open-source Chinese-Russian dictionary BKRS)