Global liberal democracy faces a near unprecedented list of digital threats in 2024 as the increasing exploitation of AI and the rampant spread of disinformation threaten the integrity of elections in more than 60 countries. And we are woefully unprepared.
Votes are scheduled in India, Pakistan, Mexico and South Africa, to name but a few. A hotly contested election will be held for the European parliament in June and the US presidential elections are on the horizon in November. A general election is also due in the UK at some stage in the coming year.
These elections are all happening at a time when global security and the very foundations of democracy are under significant strain from the rise of populism, far-right ideologies and fascist movements. Meanwhile, trust in mainstream institutions like politicians and the media remains extremely low.
What we might have once dismissed as outlandish conspiracy theories, such as that Taylor Swift is secretly working for the Pentagon and the Super Bowl is rigged, are gaining traction, and social cohesion is fraying as people segregate into isolated echo chambers online.
There is a real danger that unless we act now to protect the public these issues will only be exacerbated by the threats posed by AI, Russian disinformation campaigns, and the invasive use of technology to target voters in the coming months.
AI, deepfakes and disinformation
It's already clear that 2024 will be known as the year of the first AI elections. AI's ability to harvest near infinite amounts of data into actionable intelligence, and produce personalised content to sway public opinion will assuredly be used by mainstream political parties seeking to gain a tactical advantage in campaigning.
We are already seeing parties use AI to analyse data on voting patterns and targeting voters in real-time with algorithmically-driven ad placements.
There's nothing inherently wrong or illegal about that, though it will alarm civil libertarians and does need to be regulated. The malevolent uses of AI by rogue actors is far more concerning. Deepfakes - false or manipulated texts, images, video and audio - are already being spread via the gaming of algorithms with the intention of manipulating voters.
A deepfake AI manipulated voice of US president Joe Biden was already deployed in New Hampshire, urging voters not to turn out in its primary contest last month. During Slovakia's parliamentary elections last year, a deepfake audio recording went viral on social media, falsely depicting a party leader claiming to have rigged the election and planning to increase beer prices.
There are allegations that deepfakes were used in an attempt to sway voters in Argentina, New Zealand and Turkey in the past year. It's certain we will see highly sophisticated deepfakes circulated in many countries by rogue actors in the coming months in an attempt to influence voters, sow dissent, and put politicians on the defensive.
Bad actors
The potential for state orchestrated disinformation campaigns is evidently also a concern in the democracies holding elections this year. US State Department officials have claimed that Russia is planning to use disinformation to try to influence public opinion against Ukraine during the numerous elections scheduled across Europe this year.
In October last year the US sent a declassified intelligence assessment to more than 100 governments accusing Moscow of using spies, social media and sympathetic media to spread disinformation and erode public faith in the integrity of election outcomes. Just last month the German Foreign Ministry disclosed that its security agencies had exposed an extensive pro-Russian disinformation operation, orchestrated using thousands of fake social media accounts.
NATO and the European Union have also warned against the threats to democratic cohesion caused by Kremlin-fuelled disinformation campaigns.
In India, the ultra-nationalist government of Narendra Modi has been accused of running a covert disinformation operation, circulating propaganda to discredit foreign critics, attack political opponents and target Muslims and other ethnic and religious minorities. Human Rights Watch reports increased attacks against ethnic and religious minorities including Muslims, as well as journalists and opposition leaders.
Taking action
Calling for action now is almost moot as it's probably already too late. The fact that there are so many elections happening simultaneously around the world in 2024 only exacerbates the problem. However, we must at least try.
An urgent global effort among nations is needed to set the ground rules for how the use of AI is to be regulated, particularly around elections. The US Senate is currently considering the Protecting Elections from Deceptive AI Act, while the EU reached a tentative agreement in December to regulate AI, becoming the first major global power to do so.
Laws need to force transparency in how AI models are trained and deployed, and require disclosure for when they are used in political campaigning. The worry is that the pace at which the technology is advancing is outpacing efforts to safeguard the public.
Social media platforms must be held accountable for disinformation spread. Companies like X, Meta and Alphabet have downsized teams dedicated to integrity, hindering proactive disinformation countermeasures. Tough new laws are needed to force these tech monoliths to tackle disinformation and force transparency in algorithms and political ad targeting.
Proactive strategies like pre-bunking (teaching people to spot fake news) and rapid response strategies are essential to combat election interference. Media outlets also need to learn from past mistakes and balance truthful reporting with free speech, avoiding the "false balance" trap of amplifying disinformation from populist politicians masquerading as legitimate discourse.
Finally, we must find ways to tackle the echo chambers and conspiracy theories that threaten to derail social cohesion. Gaining back public trust in institutions such as the mainstream media and government is not going to be easy.
There are no magic spells to fix this overnight. But we can't just sit back and accept the status quo. Education in media literacy is also vital to defend against disinformation.
But while these steps may keep the mainstream parties honest, they will do nothing to stop the bad actors. Russia, China and Iran are all likely to attempt to shape geo-political outcomes in their favour in 2024 by attempting to interfere in elections.
The stability of global democracy may well depend on how these emerging threats are navigated in the months to come. When Donald Trump claimed the 2020 election was stolen, thousands of his supporters stormed the US Capitol. He may well be the president of the US again in November.
Tom Felle, Associate Professor of Journalism, University of Galway