In a review of its content moderation efforts surrounding last year's US midterm elections, Twitter said political influence campaigns were far less active last year in comparison to the 2016 US presidential election.
Nevertheless, the social media giant was eventually forced to intervene and take down accounts "connected to sources within Iran, Venezuela, and Russia," said Carlos Monje Jr., Twitter Public Policy head.
"The majority of these accounts were proactively suspended in advance of Election Day due to the increasingly robust nature of our technology and internal tooling for identifying platform manipulation," he said.
The Twitter exec did not go into any details regarding the type of content these accounts posted.
Besides political influence campaigns that tried to sway public opinion, Twitter said it also intervened to stop attempts at voter suppression.
These attempts included tweets that tried to intimidate or deter groups of people from voting, or tried to pass incorrect information about the voting or registering to vote process. The surprising detail was that most of these tweets came from accounts in the US, and not from other countries.
"We took action on nearly 6,000 Tweets we identified as attempted voter suppression, much of which originated here in the United States," Monje Jr. said, putting a number on the company's efforts.
Ignoring these bad faith tweets, Twitter said that the vast majority of the 99 million tweets related to the US midterms were mostly positive, from "individuals expressing their views on issues and candidates they care about, and encouraging others to vote."
Data connected to the Iran, Venezuela, and Russia-linked accounts that have tried their hand at a political influence campaign has been added to an open source archive containing info about all the previous influence campaigns that have been carried out on Twitter, and which the company released to the general public and political researchers last fall.
Besides data on the accounts that tried to meddle with the US midterm elections, Twitter also added information on other accounts that engaged in political campaigns last year aimed at other countries or on non-midterm election topics.
This included accounts originating in Bangladesh (tied to state-sponsored actors operating inside the country), Iran (an additional 2,617 accounts on top of the previous 284 that Twitter banned in August), Russia (an additional 418 accounts after Twitter banned IRA-linked accounts last spring), and Venezuela (764 accounts engaging in a political influence campaign against another country, and 1,196 accounts engaged in an influence campaign against Venezuelans).
Also today, Facebook, too, removed 262 pages, 356 accounts, 3 groups on Facebook, as well as 162 accounts on Instagram that it had tied to another political influence operation originating out of Iran.
Facebook said these accounts tried to influence public opinion in countries such as Afghanistan, Albania, Algeria, Bahrain, Egypt, France, Germany, India, Indonesia, Iran, Iraq, Israel, Libya, Mexico, Morocco, Pakistan, Qatar, Saudi Arabia, Serbia, South Africa, Spain, Sudan, Syria, Tunisia, US, and Yemen.
More security coverage:
- New security flaw impacts 5G, 4G, and 3G telephony protocols
- DOJ moves to take down Joanap botnet operated by North Korean state hackers
- Company selling social media 'likes' and 'followers' settles with US authorities
- Japanese government plans to hack into citizens' IoT devices
- DailyMotion discloses credential stuffing attack
- IoT botnet used in YouTube ad fraud scheme
- California governor signs country's first IoT security law CNET
- The Japanese government plans to hack into IoT devices TechRepublic