While we are still learning what role social networks play in the electoral process and we are already witnessing misusing social networks in election campaigns, the development of new technologies continues and AI is gaining momentum all over the world.
Of particular concern is the increasingly prevalent ‘deepfake’ technology, which uses AI to create highly realistic fake images, videos or audio recordings that can show individuals saying or doing things they have never actually said/done.
Examples of how AI can be misused in election campaigns have already begun to appear in some countries. In the USA, in January 2024, a robocall that appears to be an AI voice resembling President Joe Biden is reaching out to New Hampshire residents, advising them against voting. In India, in last year’s elections in November, as voters lined up to cast their ballots in the parliamentary elections to choose the government of the southern Indian state of Telangana, a seven-second clip began to go viral on social media. Namely, the opposition Congress Party showed the leader of the state urging people to vote for the Congress Party. Also, before the September 2023 elections in Slovakia, just two days before the elections, an audio recording was published on Facebook. Two voices were heard on it: reportedly Michal Šimečka, who leads the liberal Progressive Slovak Party, and Monika Tódová from the daily newspaper Denník N. They appeared to be discussing how to rig the election by buying votes from the country’s marginalized Roma minority. A few days before that, another video surfaced in which his voice could be heard talking about increasing the price of beer.
Although AI can also have a positive use in election campaigns, it seems that it will primarily be used to discredit political opponents and to spread misinformation, that is, to change the public’s perception of certain candidates in elections or parties.
An additional problem is that such fake pictures, videos and audio recordings have a predisposition to go viral on social networks due to their sensational nature, which increases their reach and influence on citizens-voters. On the other hand, citizens-voters, due to the increasing progress of new technologies, have fewer and fewer tools to distinguish fake content from the real one.
When speaking about AI in Bosnia and Herzegovina, content created by AI can be seen on social networks. When it comes to political figures, such content is currently present just in a satirical nature.
Given that the 2024 Local Elections are approaching, and that social networks are already widely used for a similar purpose, i.e. discrediting political opponents and creating a false image of mass support, the question arises whether AI will play a similar role in the upcoming elections and if so in which way.
Considering that some countries have already begun to confront the misuse of AI in election campaigns it is not unexpected that it will be used in the same way in Bosnia and Herzegovina. That is why it is very important to timely establish regulatory measures so to regulate the misuse of AI, but also to launch more initiatives for media literacy so that citizens-voters have the tools to resist false content in the electoral context.
