The social media giant announced this morning that it will open an A.I. lab in Paris, expanding its current A.I. research team of more than 40 people in New York and Menlo Park, California. "It's our hope that this research will ultimately help us make services like News Feed, photos and search even better and enable an entirely new set of ways to connect and share," the company said in a statement.
Facebook's facial recognition software turned heads when it first arrived, as it did its newsfeed, but it's been a while since Facebook users have openly revolted against a major change to the service. Maybe that revolt is coming when the A.I. finally trickles out, or maybe we've simply surrendered. The present moment is dominated by paranoia about A.I., in general, from dramas like Ex Machina to kid-friendly films like Inside Out. And while the pervasive fear extends far beyond social networks (into, say, the connected home or the smartest computer only getting smarter), Facebook already represents the closest thing we have to Transhumanism at a mass scale.
Facebook's seamless Instagram integration allows not only our best memories but our best filtered memories to be stored in the cloud in perpetuity. Accounts don't easily go away after we die. Most of us no longer keep physical address books, but we know what cities and towns our Facebook friends live in. None of these observations are groundbreaking, but when it comes strictly to A.I., Facebook has a clear advantage over other developers: Facebook knows how we think both as humans (emotional impact, as measured by likes and shares) and as machines (location queries, time on site, number of refreshes per day). Facebook, better than anyone, knows how we ingest a new piece of stimuli and how we react to it. Facebook knows what keywords and emoticons we use, both on a micro and macro scale. Facebook knows that Jimmy Fallon and John Oliver reliably hold our attention. Facebook has direct, quantifiable metrics about various human emotions. Facebook knows that people like boobs.
When the above facts are taken together, and when the accompanying research is soon improved upon by Silicon Valley minds (with Silicon Valley salaries), Facebook's A.I. game seems less about what do they want to do and more about, well, how much do they want to do?