Abigail Darwish | 18 September 2024

Summary

• Ahead of the US presidential elections in November, Russia has progressively employed  AI to influence electoral results.

• Election interference has so far targeted particular regions and demographics in America in order to influence and manipulate voter opinions. 

• Concerted efforts by the US Justice and Treasury Departments and non-governmental entities such as Meta, have undertaken countermeasures to Russian interference.

As the 2024 US presidential election looms, Russia’s leveraging of artificial intelligence (AI) to deploy disinformation, manipulate public opinion, and influence the electoral outcome is set to intensify. Though Russia’s interference in foreign elections is not unprecedented – its successful meddling in the 2016 US election is one of many instances – their renewed efforts are of equal, if not greater concern. According to a senior official at the Office of the Director of National Intelligence, Russia’s efforts are now “more sophisticated than in prior election cycles.” 

This escalation is likely due to significant advances in AI over the past decade, making it an even greater threat to the integrity of democratic processes. Simultaneously, broader geopolitical dynamics, notably the increasing erosion of relations between Russia and the West, have further incentivised foreign interference. US-Russian relations are particularly strained this electoral cycle with American support for Ukraine in the ongoing  Russo-Ukrainian  War. Correspondingly, AI has been leveraged not only to undermine democracy in America and the West more generally but also to curb Western support for Ukraine ahead of the elections. With hindsight of Russia’s successful year of cyber interference in 2016, American intelligence and governmental departments have been significantly more proactive in countering these threats. 

Russian Efforts and US Countermeasures

One feature of Russia’s interference strategy has been deploying the ‘Doppelgänger’ cyber campaign. Operated by the Russian Social Design Agency and Structura National Technologies since May 2022, Doppelgänger seeks to disseminate both disinformation and pro-Russian propaganda online and ultimately manipulate voter opinions. Using generative AI, Doppelgänger creates content and acquires ‘cybersquatter’ domain names to appear as authentic media to imitate news outlets, think tanks and even governmental organisations. To circulate such fabricated information online and foster some air of legitimacy, generative AI has been used to create social media profiles posing as US citizens to comment and reshare the materials.  

In response to this, the US Justice Department recently seized 32 Russian-based internet domains used by the Russian Government and Russian Government-sponsored actors as part of the Doppelgänger campaign. According to this indictment, FBI agents discovered six articles published on a Doppelgänger website, washingtonpost[.]pm. One article titled “White House Miscalculated: Conflict with Ukraine Strengthens Russia” purported that the US Government should recognise that continued support for Ukraine is a “mistake”, “a waste of lives and money” and “only means further destruction”, and it is therefore in the interests of all parties involved that the Biden Administration should “just make a peace agreement and move.” Though this is just one example, it illustrates how AI has been used not only to advance pro-Russian opinion through malicious means, namely posing as a legitimate US media outlet but also to disparage the Democrat Party’s foreign policy. The latter is hardly surprising given Putin’s favourability towards  Trump, and the more conciliatory positions adopted toward Russia by both Trump and JD Vance amidst the Russo-Ukrainian War. 

Alongside the efforts of the US Dept. of Justice, the US Treasury has endeavoured to address and “aggressively counter” further Russian AI operations. The Treasury has recently made public that Russian state-sponsored actors have used generative AI, specifically deep fakes, in order "to undermine confidence in the United States’ election process and institutions." The Treasury has also identified a Russian non-profit organisation, Autonomous Non-Profit Organisation Dialog (ANO Dialog), founded by the Moscow city government in 2019, as another avenue through which Russia “leverages AI technology” to disseminate “online Russian disinformation”, specifically in electoral campaigns. In response, amongst 10 individuals and two entities, ANO Dialog was ‘designated’ by the Treasury’s Office of Foreign Assets Control, meaning that those parties targeted have their assets blocked and US persons are prohibited from dealings with them, as a means of countering Russia’s strategy of interference in the elections. 

Concerningly, in the wider context of Russian interference, these efforts to coerce public opinion have not been arbitrarily disseminated online, but rather strategically aimed at specific regions and demographics, according to Deputy Attorney General Lisa Monaco. Using AI and influence-for-hire firms, residents of swing states– crucial to the overall electoral outcome – as well as American Jews, US citizens of Hispanic descent, American gamers, and users of the Reddit and 4chan online communities.

However, non-governmental entities have also assisted in efforts to counter Russian interference. According to a recent security report by Meta, the company has successfully disrupted at least 39 Russian influence operations. Of these covert operations, many have employed generative AI to create fake journalist personas and upload fabricated and divisive stories on fake news sites.

Forecast

• Short-term

  • It is very likely that despite US efforts Russian interference will continue, if not intensify, as the elections approach. 

  • It is likely that Russia’s use of AI will continue to favour Trump, which aligns more generally with Russian strategy in previous elections cycles (i.e., 2016 and 2020), and its contemporary objective of undermining Western support for Ukraine in the ongoing Russia-Ukraine War.

• Long-term

  • It is likely that public distrust in democratic processes will continue to erode, subject to the US Government’s ability to curtail Russian interference.

  • The trajectory of securitisation of social media platforms will likely persist as a necessary measure to curb foreign interference in future election cycles, as companies such as Meta have been doing.