Facebook Makes Elusive Vow to Rethink User Experimentation The social network will give its researchers 'clearer guidelines' for future experimentations -- but just what these guidelines are is anyone's guess.
By Geoff Weiss
Opinions expressed by Entrepreneur contributors are their own.
Three months after a Facebook research experiment secretively manipulated the emotional coloring of 700,000 user feeds -- much to the shock and disgust of Facebookers across the globe -- the company said today that it is changing the way it will approach future studies.
But in the same breath that the company's chief technology officer, Mike Schroepfer, described having "taken to heart" criticism of the psych experiment and vowed a more "responsible" framework for research going forward, he offered little insight into what the company's reimagined policies will look like.
For instance, Facebook said it would give its researchers "clearer guidelines" -- especially if the research "relates to content that may be considered deeply personal." However, it made no mention of what these guidelines are.
Related: Seeking Truce With Drag Community, Facebook Will Amend Its 'Real-Name' Policy
Schroepfer also said that Facebook has created a review panel comprising senior members of its engineering, research, legal, privacy and policy teams in order to safeguard the ethics of future experiments. However, notes The New York Times, no outside body will be privy to these discussions.
Additionally, Facebook said it has updated its six-week training program -- or "bootcamp," as it's known -- to better educate researchers about privacy. And going forward, the company will publish all academic research in a single location.
Of particular note, however, was that Facebook neglected to indicate whether it would seek user consent prior to conducting future experiments. "In essence," writes the Times, "Facebook's message is the same as it has always been: Trust us, we promise to do better."
Related: OkCupid Founder: 'If You Use the Internet, You're the Subject of Hundreds of Experiments'