Deepfakes & Elections; Is Asia Prepared for Misinformation Campaigns?

Before the Indonesian elections on February 14th, a video featuring the late Indonesian president Suharto endorsing the political party he once led gained immense traction online. The video, created using AI technology to mimic his appearance and voice, garnered 4.7 million views on X alone. However, this wasn’t an isolated incident.

In Pakistan, a deepfake video surfaced during the national elections, purportedly showing former prime minister Imran Khan announcing his party’s boycott of the elections. Similarly, in the U.S., voters in New Hampshire heard a deepfake of President Joe Biden urging them not to vote in the presidential primary.

Such instances of deepfakes involving politicians are becoming more frequent, especially with 2024 poised to be a significant global election year. With approximately 60 countries and over four billion people set to cast their votes, the proliferation of deepfakes raises serious concerns.

According to a November report by Sumsub, the global incidence of deepfakes increased tenfold from 2022 to 2023. In the Asia-Pacific region alone, there was a staggering 1,530% surge in deepfake incidents during the same period.

Various sectors, including online media, social platforms, digital advertising, and professional services, witnessed a substantial rise in identity fraud rates between 2021 and 2023.

Simon Chesterman, senior director of AI governance at AI Singapore, expressed concerns about Asia’s readiness to combat deepfakes in elections, citing deficiencies in regulation, technology, and education.

Crowdstrike’s 2024 Global Threat Report highlighted the likelihood of nation-state actors, including those from China, Russia, and Iran, conducting misinformation campaigns during this year’s elections.

While some governments have tools to counter online falsehoods, there’s apprehension that the spread of deepfakes may outpace regulatory efforts.

Most deepfakes are anticipated to originate from actors within the respective countries, although foreign interference remains a significant concern.

Carol Soon, principal research fellow at the Institute of Policy Studies in Singapore, identified potential domestic actors such as opposition parties, political opponents, and extremist groups.

The proliferation of deepfakes poses various dangers, including polluting the information ecosystem, influencing voter perceptions, and eroding trust in electoral processes.

There’s a growing realization of the need for social media platforms to assume greater responsibility due to their influential role in disseminating information. In February, leading tech companies announced a joint commitment to combat the deceptive use of AI in elections.

However, implementing and enforcing these measures effectively remains a challenge, requiring transparent decision-making processes and multi-pronged approaches.

Chesterman stressed the importance of establishing regulations and setting expectations for tech companies, given the limitations of relying solely on their goodwill.

Efforts like the Coalition for Content Provenance and Authenticity (C2PA) aim to enhance content verification through digital credentials, with member companies including Adobe, Microsoft, Google, and Intel.

While technological solutions are part of the equation, consumer awareness and education are equally crucial. Vigilance, fact-checking, and responsible sharing of information are essential in combating the spread of deepfakes.

In conclusion, addressing the threat of deepfakes in elections requires collaborative efforts involving government regulation, technological innovation, and public awareness campaigns.