Speed-tweet­ing hu­man ‘bots’ are on­line

Some fall vic­tim to Twit­ter’s vote crack­down

San Antonio Express-News (Sunday) - - Nation - A S S O C I AT E D P R E S S

CHICAGO — Nina To­masieski logs on to Twit­ter be­fore the sun rises.

Seated at her din­ing room ta­ble with a nearby TV con­stantly tuned to Fox News, the 70-year-old grand­mother spends up to 14 hours a day tweet­ing the praises of Pres­i­dent Don­ald Trump and his po­lit­i­cal al­lies, par­tic­u­larly those on the bal­lot this fall, and de­rid­ing their op­po­nents.

She’s part of a ded­i­cated band of Trump sup­port­ers who tweet and retweet Keep Amer­ica Great mes­sages thou­sands of times a day.

While her goal is to ad­vance the agenda of a pres­i­dent she adores, she and her friends have been swept up in an ex­panded ef­fort by Twit­ter and other so­cial me­dia com­pa­nies to crack down on ne­far­i­ous tac­tics used to med­dle in the 2016 elec­tion.

And with­out mean­ing to, the tweet­ers have demon­strated the dif­fi­culty such crack­downs face — par­tic­u­larly when it comes to telling a po­lit­i­cal die-hard from a sur­rep­ti­tious com­puter ro­bot.

Face­book re­cently said it had re­moved 32 fake ac­counts ap­par­ently cre­ated to ma­nip­u­late U.S. pol­i­tics, ef­forts that may be linked to Rus­sia.

Twit­ter and other sites also have tar­geted au­to­mated or robot­like ac­counts known as bots, which au­thor­i­ties say were used to cloak ef­forts by for­eign gov­ern­ments and po­lit­i­cal bad ac­tors in the 2016 elec­tions.

But the screen­ing re­peat­edly and er­ro­neously has flagged To­masieski and users like her.

Their ac­counts have been sus­pended or frozen for “suspi- cious” be­hav­ior, ap­par­ently the fre­quency and re­lent­less­ness of their mes­sages.

When they started tweet­ing sup­port for a con­ser­va­tive law­maker in the GOP pri­mary for Illi­nois gov­er­nor this spring, news sto­ries warned that rightwing “pro­pa­ganda bots” were try­ing to in­flu­ence the elec­tion.

“Al­most all of us are con­sid­ered a bot,” said To­masieski, who lives in Ten­nessee but is tweet­ing for GOP can­di­dates across the U.S.

The ac­tions have drawn crit­i­cism from con­ser­va­tives, who’ve ac­cused Twit­ter, Face­book and other com­pa­nies of lib­eral bias and cen­sor­ship.

It also raises a ques­tion: Can the com­pa­nies out­smart the ever-evolv­ing tac­tics of U.S. ad­ver­saries if they can’t be sure who’s a ro­bot and who’s Nina?

“It’s go­ing to take a re­ally long time, I think years, be­fore Twit­ter and Face­book and other plat­forms are able to deal with a lot of these is­sues,” said Ti­mothy Carone, who teaches tech­nol­ogy at Notre Dame’s Men­doza Col­lege of Business.

Twit­ter didn’t re­spond to a re­quest for com­ment. But the com­pany has said it iden­ti­fied and challenged close to 10 mil­lion sus­pected bot or spam ac­counts in May, up from 3.2 mil­lion in Septem­ber.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.