What Would I Say?
"What Would I Say?" is one of the hottest new apps to hit Twitter and Facebook in a while, and for good reason: it's hilarious. What Would I Say?

"What Would I Say?" is probably popping up all over your Facebook and Twitter feeds, as it's one of the hottest new games to hit social media in recent memory.

The app is based on a simple premise. Users allow "What Would I Say?" access to their Facebook accounts, then they are presented with two buttons: an orange one reading "Generate Status" and a blue one stating "Post To Facebook."

When a logged-in Facebook user clicks the orange button, the site instantly creates a status based on their past statuses, often cranking out hilarious sentences, and other times coming up with ridiculous strings of unrelated words that can be funny in their own ways. Whichever way it goes, the whole thing is based on you, which may explain the app going viral over the past day.

And it's pretty much brand-new, as it came from the great minds assembled over this weekend at HackPrinceton 2013.

It's mostly just a fun diversion, but it's based on fairly complicated computer science, as the "About" section of the "What Would I Say?" website explains.

"'What would i say?' automatically generates Facebook posts that sound like you!" it explains. "Technically speaking, it trains a Markov Bot based on mixture model of bigram and unigram probabilities derived from your past post history."

Markov bots, named for programmer Andrey Markov, essentially exist to randomize words or other data points in a specific manner. By applying them to past Facebook statuses you can get some pretty choice results, as the site notes. Here's a few of their favorite "What Would I Say?" posts:

"These ubiquitous creatures are recruited using scholarships."

"President Eisgruber greets the president on behalf of an EKG that includes black squirrels."

"Save the environment, cancel quals."

"Why you should never touch raw chicken with electricity."

And just in case you thought the people behind the site were heartless robots themselves, the homepage provides a link for users to donate to the relief effort for people affected by Typhoon Yolanda/Haiyan in the Philippines.

Now that you know what the app is and have a basic concept of how it works, all that's left to do is check it out for yourself!