A competition to sneak computer-controlled "socialbots" into groups of people on Twitter has been surprisingly successful, according to New Scientist.
Organised by Tim Hwang, director of the Web Ecology Project, the game saw three teams create a Twitter bot to fool as many real people as possible.
The winning team's bot gained 107 followers and almost 200 responses to its messages in 15 days.
To generate convincing responses, the bots analysed the updates of people in the target group they were trying to infiltrate and mimicked them.
Sound a bit creepy? Wait, it gets worse.
Since the competition ended last month, the Web Ecology Project has made the source code of the bots available for download.
That means anyone can now run their own versions of the bots on Twitter — and chances are, you wouldn't be able to pick them.
All this may sound sinister, but Mr Hwang was upbeat about what the bots could offer humanity.
"We could use these bots in the future to encourage social participation or support for humanitarian causes," he told New Scientist.
On his blog, Mr Hwang said one of the biggest successes of the competition had been bridging the gaps between real Twitter users.
"In two weeks, the bots were able to heavily shape and distort the structure of the network," he said.
"This included bringing people together not originally connected, and bringing together a community of activity around the bots themselves."
In other words, the bots were like a mutual friend that brought other, human, Twitter users together.
To prove his theory on a wider scale, Mr Hwang has announced another, far more ambitious project called "The Narrows".
The challenge? To bring together two completely unrelated groups of 5000 Twitter users, without their knowledge, using "waves" of bots running for as long as 12 months.
If you're interested in taking part in the project, applications are now open.
Everyone else might want to keep an eye on any chatty new friends.

No comments:
Post a Comment
Let leave your comment here, thanks for visiting ...