I am sure IBM will pay couple of millions every year for that kind of data.

Google can also pay for it, they might use for Google translate. Wikipedia already has near similar articles in major languages. Google can fight a war for this much of user translated/verified content.