Bots, good or bad, dominate Twitter conversation: study

Twitter logos on a computer screen in Beijing: some two thirds of all popular links pushed by the social media platform were gen
Twitter logos on a computer screen in Beijing: some two thirds of all popular links pushed by the social media platform were generated by bots, a study said

Automated accounts or "bots" play a big role in disseminating information on Twitter, accounting for two-thirds of tweets linking to popular websites, a study showed Monday.

The Pew Research Center report found bots were a major source for diffusing information on news, sports, entertainment and other topics.

The researchers found that of all tweeted links to popular websites, 66 percent were shared by accounts that appeared to be automated rather than human users.

While bots have gained attention due to concerns over Russian-sponsored manipulation of social media during the 2016 political campaign and for other hot-button topics, the researchers said they made no effort to distinguish between "good" or "bad" bots.

"The study does not find evidence that automated accounts currently have a liberal or conservative 'political bias' in their overall link-sharing behavior," the researchers wrote.

Twitter's policy on automated accounts, last updated in November, allows bots to operate but with limitations.

The policy allows for bots to "automatically broadcast helpful information" or "run creative campaigns that auto-reply to users."

But Twitter's rules forbid automatic posts about trending topics or using automation "to attempt to influence or manipulate trending topics." It also bans the use of multiple accounts to generate more activity.

"These findings illustrate the extent to which bots play a prominent and pervasive role in the social media environment," says Pew researcher Aaron Smith.

"Automated accounts are far from a niche phenomenon: They share a significant portion of tweeted links to even the most prominent and mainstream publications and online outlets. Since these accounts can impact the information people see on social media, it is important to have a sense their overall prevalence on social media."

Pew researchers said some examples of automated accounts included a Netflix bot which tweets when new shows are added to the online streaming service, and another which posts random images from the Metropolitan Museum of Art.

CNN operates a "breaking news" bot and the New York Times has one that provides live analysis of NFL games.

The researchers said a small number—some 500—of highly active bots are responsible for a disproportionate share of the tweeted links, or 22 percent.

Pew found that an unusually large share of "adult content" is retweeted by bots, accounting for 90 percent of links to popular adult sites.

That coincides with findings from other researchers on campaigns of "pornbots" which advertise sex and pornographic websites.

The Pew report examined some 1.2 million English language tweets linking to 2,315 of the most popular websites in a six-week period in mid-2017.

Explore further: How do you spot a Russian bot? Answer goes beyond Kremlin watching, new research finds