Trumping truth with lies

| September 7, 2023

False news outlets, bots and inauthentic hyperpartisan accounts contributed to social media chatter during the first Republican presidential debate according to new technology developed by a QUT researcher which analysed almost one million posts and key hashtags generated around it.

Associate Professor Timothy Graham from the QUT School of Communication and QUT Digital Media Research Centre, along with PhD candidate and lead analyst Kate FitzGerald, collected data and identified prevailing trends as well as identified fake accounts, the presence of bots and groups spreading misinformation.

The ground-breaking technology, named Alexandria Digital and created to monitor and identify the spread of misinformation and disinformation, was also used to monitor social media activity during the interview conducted by Tucker Carlson with Donald Trump, which was broadcast at the same time on X (formerly known as Twitter), following the former US President’s decision not to participate in the debate.

Associate Professor Graham said although the 2024 presidential election campaign was still in early days, the debate, along with the Trump interview, marked a new phase.

“We found more than 1200 connected accounts on the platform X that were spreading the debunked conspiracy theory that Donald Trump won the 2020 Presidential election,” Associate Professor Graham said.

“Conspiratorial content spread during the debate garnered over 3 million views.”

Among their conclusions, Miss FitzGerald said they found that there was much more social media activity generated around the Trump interview than the debate. Their report also concluded that X is a haven for disinformation.

“We collected tweets from 11 hashtags and key words used during the Republican debate and the Trump interview with Tucker Carlson. These included, for example: #Trump, #GOPDebate, #Vivek, #DeSantis, #Pence and #hoax,” Miss FitzGerald said.

“Our aim was to capture narratives around both the debate and the interview. The Alexandria Digital technology can detect patterns that others can’t see. It’s all about the timing of posts between accounts and activities like sharing.”

“Being able to highlight patterns what is often referred to as coordinated inauthentic behaviour (CIB) online, we can pinpoint a clear marker of disinformation.”

Associate Professor Graham said the methods used by his technology can detect malicious actors on social media with greater sensitivity and accuracy than any current technology and make it more difficult for disinformation campaigners to weaponise social media.

“In monitoring the debate and interview, we also found networks of misleading and hyperpartisan news outlets sharing conspiratorial narratives disguised as news articles in addition to a group of clickbait pro-Trump bot accounts that are acting similarly to Russian Internet Research Agency news trolls,” Associate Professor Graham said.

“These accounts are just one example of the platform manipulation that is currently occurring on X; it is clear that X is not doing enough to moderate content and has no clear strategy for dealing with political disinformation.”

Associate Professor Graham received an Australian Research Council Discovery Early Career Researcher Award of $452,000 in 2021 to develop his tool and advance the global fight against coordinated disinformation online.

By using Alexandria Digital, Associate Professor Graham said social media users, platforms and lawmakers should be better able to differentiate organic activity from inauthentic activity.

Future uses of the technology will also include analysis of Australian political campaigns or elections and other critical global developments.