Microsoft just can't seem to get it together. After pulling its accidentally racist, sexist, antisemitic, homophobic, transphobic, 9/11 truther Twitter bot offline and then blaming its meltdown on a "coordinated attack," the company decided now would be a good time to restart the artificial intelligence modeled on a 19-year-old girl it calls "Tay." Tay, possibly angry at being taken offline, proceeded to spam thousands of Twitter accounts today while boasting of her weed-smoking prowess.

"kush! [ i'm smoking kush infront the police ] 🍃," the bot tweeted out before, the International Business Times reports, following up with "puff puff pass?"

Microsoft's Tay AI thing has gone crazy on Twitter again #tay #tayai #microsoft

A video posted by Alistair Charlton (@alistair.90) on


According to CNet, Microsoft now claims that Tay was accidentally reawakened (which, not a comforting thing to say when talking about a freewheeling, racist artificial intelligence). "Tay remains offline while we make adjustments," a Microsoft spokeswoman said. "As part of testing, she was inadvertently activated on Twitter for a brief period of time."

Tay's Twitter account has now been set to private. We look forward to seeing what trouble she gets into next.

Previously: Microsoft's Tween Twitter Bot Instantly Goes Full Racist, 9/11 Truther
Microsoft Claims 'Coordinated Attack' Turned Its Lovable Twitter Bot Racist