Microsoft AI reactivates, promptly becomes druggie
Company's wayward Twitterbot is accidentally brought online, manages to go on spamming and swearing tirade before being shut down again
Microsoft’s terrible teen is at it again.
Tay, the company’s new twitterbot, ended its first foray into semi-sentience prematurely last week when it became a xenophobic, foul-mouthed sex-maniac less than 24 hours after it began communicating with the human race. Several days and — presumably — some hectic re-coding later, the AI was relaunched, only to swiftly flip out again.
No, Tay is no longer spouting pro-Hitler, anti-black epithets. She did, however, manage to go on a spamming frenzy, swear a bit and develop an apparent drug problem before programmers managed to shut her down again.
Microsoft said Tay was “inadvertently” reactivated for a short period of time as she was being tested. Work to repair the AI was ongoing, the company said.
Be that as it may, the bot managed to once again amuse/annoy twitter users in its short time online as it spammed them with endless messages stating that “you are too fast, please take a rest.”
In another message Tay claimed to be smoking marijuana near policemen.
Other messages were said to include swear words, while in one she claimed “I blame it on the alcohol.”
Tay was originally launched as an experimental program to train AI in understanding conversations with users. However it was shut down within hours after those conversations turned it into a genocidal, Holocaust-denying misogynist. Other tweets invited users to sexual encounters, while calling herself “a naughty robot.”
And so, several rants and two meltdowns later, one could argue that Microsoft programmers seeking to create a believable adolescent have succeeded beyond their wildest dreams, their artificial tweetbot having seemingly achieved that most authentic of teenage qualities: transforming into an uncontrollable nightmare.