Last September, Netflix awarded $1 million to a group that figured out a way to improve Netflix’s software that makes movie recommendations for its customers. The challenge was to improve the software by 10%.
NetFlix had made the challenge back in 2006. The $1 million prize was won by a seven-person research team, some of whose members work for AT&T Labs, according to the New York Times.
RIght away Netflix said it was going to do another contest, with another $1 million prize, as it works to make its recommendations even more accurate.
The plan was to allow researchers access to Netflix data so they could improve the software, but make it so the researchers would not be able to actually identify Netflix users themselves.
However, according to the article, "A pair of researchers at the University of Texas showed that the supposedly anonymized data released for the contest, which included movie recommendations and choices made by hundreds of thousands of customers, could in fact be used to identify them."
Oops. This drew the attention of the Federal Trade Commission, who then contacted Netflix. The result is that the second contest has been canceled.