Australian Outlook

In this section

Elon Musk, Twitter, and the Role of Free Speech in International Social Media

28 Feb 2023
By Dr Gianluca Demartini
Twitter app icon on smartphone screen. Source: Yuri Samoilov/http://bit.ly/3mbvw1W

With the acquisition of Twitter, Elon Musk has introduced significant changes to the company, its products, and operations. What does the future of free speech on social media look like?

Twitter was acquired by Elon Musk in October of last year and was made into a private company. Since then, the platform has undergone significant changes, mainly focussed on ensuring the company can be profitable in the short term. While the early general understanding was that Musk would make the platform the place for anyone to say anything, as time went on, things started to look somehow different.

First, a number of staff have been made redundant to reduce costs. However, there has been a focus on retaining committed employees who can devote their time to the company and product development while laying off the others who cannot. This has certainly reduced and limited diversity in staff and poses obvious risks leading to the development of technology and products which are biased and that reinforce the views of very few people (that is, those who look and think like Musk).

The new ownership has introduced many changes to the platform itself, with many non-revenue-generating products being shut down. One very negative change that was recently introduced is in the Twitter API product. This is the technology that allows researchers and other companies to access tweets and other data automatically. This technology has always had some level of free access, but recently costly pricing schemes have been introduced and then cancelled within a short timeframe, showing the need for a clear vision. Because of this, much of academic research aiming to understand society may not be possible anymore.

On the other hand, one lucky exception is Community Notes: a crowdsourcing effort to identify and flag misinformation on Twitter. Rather than being shut down, since Musk’s takeover, the project has been revamped and extended by surfacing crowd-generated misinformation labels directly to users onto their Twitter feeds. This may be due to the fact that such annotations are shown to be of a quality comparable to that obtained by expert fact-checkers.

With the introduction of paid verified accounts, Musk is trying to diversify the sources of revenue. This is at the same time as ad revenue is drying up due in part to the changes introduced by Musk since the acquisition. The move towards being a subscription-based platform may create barriers for certain groups of users to join, making it an even stronger echo chamber with less diverse content. This is likely to result in and amplify extremist opinions. This approach of tier-1 paying users and tier-2 less powerful users is being followed by Twitter’s competing products from Meta.

Given Musk’s approach to the platform, where who pays is able to gain more visibility, we now see an approach in which the few who can afford it are allowed increase their visibility, reach, virality, and impact on the social media platform. This is obviously not good news for free speech.

What is the role of free speech in international social media?

All popular social media platforms, including others like TikTok and Facebook, have content moderation processes in place. Typically, these are human-in-the-loop systems that first make use of algorithms and computers to process the very large amounts of data being continuously posted by their users to flag content that potentially violates their guidelines. This flagged content is then sent to human content moderators (Meta make use of around 50,000 of them) to make final decisions about the content and whether it needs to be removed.

While content moderation guidelines set the tone for what content should be allowed on the platform, some content and aspects like disinformation, for example, may be very subjective. This might end up being assessed differently by different algorithms and different human moderators who thus have to make instinctive decisions reflecting their own values and beliefs.

In addition, the possibility of having paying users who may have extra allowance to post certain types of content may lead to a very unbalanced view on these platforms. The risk is also that this may happen in a very non-transparent way where the average user is not aware of being exposed to content coming from a certain (paying) political faction only.

Countries and governments that do not like the idea of freedom of speech could move towards banning access to those platforms which allow their users to say “too much” and to be critical of their political leaders.

Meanwhile, be warned, the Twitter account @realDonaldTrump, after being suspended following the January 2021 riots and although still silent, is active again!

Dr Gianluca Demartini is an Associate Professor in Data Science at the University of Queensland, School of Information Technology and Electrical Engineering. His main research interests are Information Retrieval, Semantic Web, and Human Computation. His research has been supported by the Australian Research Council (ARC), the UK Engineering and Physical Sciences Research Council (EPSRC), the EU H2020 framework program, Facebook, and Google. He has published more than 200 scientific publications in Computer Science and received multiple Best Paper awards.

This article is published under a Creative Commons Licence and may be republished with attribution.