Assembly polls: Twitter puts in place tech, policies to counter false info

With five states heading into election, two of them as early as March 27, microblogging social media firm Twitter has upped its game. Past experience, both in India and outside, has proved the platform to be a fertile ground for spreading misinformation and disinformation, which, in an election scenario particularly, can cause harm in real life. Twitter, which has around 17.5 million users in India (according to government data), hopes its new policies, an expanded team and additional technologies will prevent – or at least limit – this.

"In line with our Civic Integrity Policy, we will take down content containing false or misleading information about (1) process and procedures, (2) false information intended to intimidate or dissuade people from participating in the elections and voting, and (3) accounts that misrepresent their affiliation with a political candidate or party," said Kathleen Reen, senior director, Public Policy and Philanthropy, APAC, Twitter, in an emailed response to Business Standard. 

The heightened social media activity ahead of an election, whether local or national, has been of concern for governments across the world. From amplifying false information to attacking political figures, and even effecting changes in election outcomes, social media platforms are being used for all.

Election-specific event page

Specific to the elections in Assam, Kerala, Tamil Nadu, West Bengal and Puducherry, Twitter will launch an event’s page on voting days and for the election results day. 

The page will include a timeline of Tweets from credible accounts to provide latest information of voting and results. The page will be visible to account holders in India in the Explore Tab, and will provide continuous updates and context throughout the election period with videos in multiple languages from a variety of news partners, Twitter said.

The election information prompt will be available to users in six languages: Bengali, Tamil, Malayalam, Assamese, Hindi and English. 

The prompt will help users find credible information about candidate lists, voting dates, polling booths and EVM voter registration among other election-related information.

Rules impartial to politicians

Another area of concern, flagged by Indian authorities to Twitter and Facebook, is that of the de-platforming of former US president, Donald Trump, for egging on his supporters to Capitol Hill earlier this year. On this, Reen said, “No one is above the Twitter rules and we strive to enforce our policies judiciously and impartially for everyone. We assess reported tweets from world leaders against the Twitter rules, which are designed to ensure people can participate in the public conversation freely and safely." 

To prevent trending hashtags from being misused or misconstrued, Twitter will show a representative tweet, Twitter Moment or description to the top trends, Twitter India said in a blog post Wednesday. 

Keeping India’s complex political and cultural context in mind, Twitter has also expanded its team. “A global cross-functional team with local, cultural, and language expertise will run our election integrity work,” the Twitter blog post said. “Their role is to keep the service safe from attempts to manipulate the platform and content that can incite violence, abuse and threats and trigger the risk of offline harm.”

Content authenticity

The platform will also label content as being synthetic or manipulated if there is enough evidence that it, or the context in which it has been put up, is significantly altered. 

“When people attempt to retweet tweets with a synthetic and manipulated media label, they will see a prompt pointing them to credible information. These tweets won’t be algorithmically recommended by Twitter, which further reduces the visibility of misleading information,” Twitter said in the post.

The "weaponisation" of Twitter trends, or using bots to amplify hashtag content and convert it to a top trend, has been an issue of concern, especially in a political scenario. On how Twitter is tackling this, Reen said, "There are rules for trends and this means that at times, we may not allow or may temporarily prevent certain terms that violate Twitter rules from appearing in the Trends section."

Twitter’s 2019 ban on political advertising continues.

The firm further said inauthentic engagement, which includes selling/buying tweets or account metric inflation (like retweets, likes, mentions, twitter poll votes), qualify as a violation and will be acted upon. Accounts that engage in such behaviour may even invite a permanent suspension.