Social Media Disinformation Threat Escalates as Generative AI Advances

Japan’s election season has been marred by a surge in online misinformation and disinformation spreading across social media platforms, raising concerns about the integrity of the upcoming House of Representatives election.

Experts note that the problem has intensified with the evolution of generative artificial intelligence technologies, which now enable virtually anyone to create convincing fake videos and manipulated images with minimal technical expertise.

In mid-January, a doctored image featuring Prime Minister Sanae Takaichi with unnaturally altered facial features circulated widely on X (formerly Twitter). The image, which garnered over 3.6 million views, was accompanied by text claiming, “Sanae Takaichi’s appearance speaks to the darkness and terror within her heart”—a clear attempt to damage her public perception.

Similarly, a deceptively edited video clip of Centrist Reform Alliance co-leader Tetsuo Saito received more than 800,000 views. The clip showed Saito saying, “There are things more important than humans,” but deliberately omitted his immediate follow-up statement, “We don’t subscribe to such a philosophy.” The selective editing significantly distorted his message and political stance.

Another fabricated video that gained viral status portrayed opposition party leaders apparently dancing during their campaign broadcast, undermining the seriousness of their political message.

While political satire has a long-standing tradition in democratic discourse, these recent examples represent a more insidious trend: the deliberate dissemination of false information designed to deceive voters and manipulate public opinion. If left unchecked, such manipulation threatens to undermine electoral fairness—the cornerstone of Japan’s democratic process.

The structural problem lies within social media’s fundamental business model. Platforms reward content that generates high engagement regardless of accuracy or societal impact. Content creators can monetize sensational posts through advertising revenue, creating a perverse incentive to produce provocative—often false—content without consideration for the broader societal damage.

Last spring, Japan implemented the Information Distribution Platform Law, which requires major platform operators to promptly address deletion requests from individuals who claim they have been harmed by specific posts. However, the law has significant limitations: platform operators maintain sole discretion over whether to remove content, and they are not required to publicly disclose their reasoning for such decisions.

Political observers suggest that both ruling and opposition parties should consider strengthening regulations governing social media content during elections. One proposed measure involves restricting the monetization of political content during official campaign periods, potentially reducing the financial incentives behind creating inflammatory or misleading election-related content.

As Japan approaches this crucial election, voters are urged to exercise critical thinking when consuming online information. Election experts recommend verifying claims through official channels such as government election bulletins, party websites, and traditional news sources before forming opinions or making voting decisions.

The proliferation of election misinformation in Japan reflects a global challenge facing democracies in the digital age. As technology continues to advance, the ease with which deceptive content can be created and disseminated threatens to outpace regulatory frameworks designed to maintain electoral integrity.

For democracy to function effectively, voters must have access to accurate information when making decisions at the ballot box. The current situation highlights the delicate balance between free expression and protecting the democratic process from deliberate manipulation—a challenge that Japanese society must address as it prepares for the upcoming election.