Twitter Bots, Politics, and Social Activism
by Sandra Mwangi
Mentor: Prof. Jasmine McNealy; Student Editor: Sanika Mahajan
With the rise of social media activism and increased power of hashtags, Twitter is becoming a powerful weapon in shaping democracy and public opinion. On a platform where the virality of hashtags and videos is a metric of how effective a campaign or movement is, bots are thriving and driving a lot of discussions. What many Twitter users are not aware of is the number of views, likes, retweets a post gets depends largely on autonomous activity- Twitter bots.
Bots (short for robots) are algorithmically driven entities that appear to be human users. Bots are used to generate messages automatically, push forward ideas, and follow other users. Bots just like any automated machine are neither good nor bad, however they can be used to spread misinformation, infiltrate groups of people and propagate specific ideas. Social bots mimic human behaviour like but should not be confused with trolls. Trolls are accounts run by humans with an agenda to advocate for or against certain topics. Usually one troll is surrounded by many bots such that when a bot’s argument gathers a lot of attention, it is passed onto a troll. Most researchers focus on Twitter because the platform allows automated accounts and politics is the most prominent topic bots are involved in.
A study by the Pew Research Center estimates that between 9% and 15% of all Twitter accounts are automated, 66% of all tweeted links to popular sites were shared by bot accounts, and 89% of links to news-aggregation sites were bot sourced. There are ways to identify a bot, though they are becoming more difficult to identify as social networks become more sophisticated. Bots tweet way more than a human can- over 50,000 tweets a day, often have alphanumeric usernames, no profile pictures or pictures of other people, no friends but lots of retweets. Because of their sheer numbers and high activity, bots wield a lot of power and confluence.
A study done by Alessandro Bessi and Emilio Ferrara of the USC Information Science Institute estimated that about 400,000 bots were engaged in the political discussion about the US 2016 Presidential election, responsible for roughly 3.8 million tweets, about one-fifth of the entire conversation. In October 2018, Twitter admitted that more than 50,000 Russian-based accounts suspected to be bots posted automated material on the 2016 US elections in support of Trump. The accounts tweeted about Trump 10 more times than they did about Hillary and were responsible for 4.25% of all retweets of Trump’s account in the last weeks of the election campaign. Even with all this, there is no clear metric to determine whether the bots skewered voters’ decisions but it is clear that the bots affected the public opinion of each candidate especially to less educated and knowledgeable users.
Across the Pacific Ocean in Africa, a study done by Portland communications on ‘How Africa Tweets’ found that 10 elections across 9 African nations in 2017–2018 have been influenced more by outside factors such as bots than by actual politicians, news media, bloggers, government, campaigners and think tanks. The study found that 53% of the leading and influential voices on Twitter came from automated activity outside the country with elections. Bots were active across all elections, most prevalently in Kenya, where they accounted for over 25% of all influential accounts in both elections- 25% for the August election and 28% for the October election.
Kenyans on twitter(KoT) are very active as Twitter is a means for the citizens to challenge their government and receive information in real time. In her book, ‘Digital Democracy, Analogue Politics: How the Internet Era is Transforming Kenya (African Arguments)’ author Nanjala Nyabola offers insights on how digital developments are shaping politics in contemporary Kenya. She writes that “For Kenyans, social media is not simply a space to post pictures of new clothes or delicious food, or to have conversations about sports. It is a space where some of the most exuberant and insightful political conversations are happening. The citizens find this especially refreshing because other forms of digital media provide government sanitized news.”
In the 2017 Kenyan elections, the Twitter bots served to propagate propaganda, push negative narratives about electoral processes and perpetuate misinformation on perceived election irregularities. The online conversation served not to promote democracy but to create a wider divide in an already tribally polarised society. In addition to social bots, data mining and psychometric profiling is another means algorithms were used to alter public perception of political candidates. Cambridge Analytica, a data company at the centre of the controversy of using personal data from psychometric profiling techniques to influence the Brexit vote in the UK, the 2016 US election results and the 2017 Kenyan elections that were later nullified by The Supreme Court. Cambridge Analytica uses voters’ information to target them with information and sway their political opinions.
Twitter has put out statements on how they are trying to curb activity of bots. There is much they can do but a lot of work has to be done by Twitter users. Users have to be aware of the presence of bots, how to identify one and what to do when you suspect an account is a bot- report the suspected account.
Internationally, Twitter needs to do better. Twitter has been under fire for prioritizing American twitter and not working on international twitter. There are no Twitter offices in Africa, despite Africa having over 2000 languages. In an interview, Larry Madowo, a prominent Kenyan journalist, says, “Although Twitter is making deliberate steps to combat bots in the United States and Europe partly because of strong regulators, bots have poisoned the political discourse in Africa, where the firm pays little to no attention.”
Oftentimes social media in politics is considered in a Western context and ignored in the international context. Twitter users all over the world should be aware of the presence of bots and how their activity influences political and social discussions. Twitter users should also closely examine any activity that influences their political decision in any way and find the source of the trending hashtag or retweet because many a time it will be a Twitter bot.
Sources
- Ndlela, Martin N. “Social Media Algorithms, Bots and Elections in Africa.” Social Media and Elections in Africa, Volume 1, 2020, pp. 13–37
- Howard, Philip N. “IEEE Spectrum.” How Political Campaigns Weaponize Social Media Bots, 18 Oct.2018,https://spectrum.ieee.org/computing/software/how-political-campaigns-weaponize-social-media-bots. Accessed 11 Aug. 2020
- Guilbeault, Douglas. and Woolley, Samuel. “The Atlantic.” How Twitter Bots Are Shaping the Election. 1 Nov 2016. https://www.theatlantic.com/technology/archive/2016/11/election-bots/506072/. Accessed 11 Aug. 2020
- Swaine, Jon. “The Guardian.” Twitter admits farmore Russian bots posted on election than it had previously disclosed. 20 Jan 2018. https://www.theguardian.com/technology/2018/jan/19/twitter-admits-far-more-russian-bots-posted-on-election-than-it-had-disclosed Accessed 12 Aug 2020
- O’Conner, Gabe. “ npr All Tech Considered.” How Russian Twitter Bots Pumped Out Fake News During The 2016 Election. 3 Apr 2017. https://www.npr.org/sections/alltechconsidered/2017/04/03/522503844/how-russian-twitter-bots-pumped-out-fake-news-during-the-2016-election Accessed 12 Aug 2020
- Twitter Inc. Update: Russian interference in the 2016 US presidential election. 28 Sept 2017. https://blog.twitter.com/en_us/topics/company/2017/Update-Russian-Interference-in-2016--Election-Bots-and-Misinformation.html Accessed 11 Aug 2020
- Roth, Yoel. and Harvey, Del. Twitter Inc. How Twitter is fighting spam and malicious automation. 26 Jun 2018. https://blog.twitter.com/en_us/topics/company/2018/how-twitter-is-fighting-spam-and-malicious-automation.html Accessed 13 Aug 2020
- Portland Communications, How Africa Tweets. 4th Edition. 2018 pp.13 https://portland-communications.com/pdf/How-Africa-Tweets-2018.pdf. Accessed 11 Aug 2020
- Bessi, Alessandro. and Ferrara, Emilio. “Social Bots Distort the 2016 US Presidential elections online discussion.” https://firstmonday.org/article/view/7090/5653 Accessed Aug 11 2020
- Nyabola, Nanjala. Digital Democracy, Analogue Politics ‘HOW THE INTERNET ERA IS TRANSFORMING KENYA’ Zed Books, Vol. 1 chapter 3–5 https://press.uchicago.edu/ucp/books/book/distributed/D/bo31262836.html Accessed 12 Aug 2020
