Sign in Subscribe
Political parties

AI-Generated Political Videos Blur Truth and Fiction in Bangladesh's Upcoming Election

Strategic Insights

— 2 min read

Bangladesh is on the cusp of a worrying phenomenon: AI-generated political videos that make it increasingly difficult to distinguish truth from fiction. These synthetic clips, featuring entirely fictional people, have gained significant traction on social media, particularly among the parties running in the upcoming general election. With AI tools like HeyGen, DeepFaceLab, and Synthesia becoming more accessible, anyone with a smartphone can create videos that convincingly mimic real people. The surge in such content, analyzed by independent online verification and research platform Dismislab.com, has garnered over 23 million views since mid-June 2025.

Key Takeaways:

  • The Bangladesh Election Commission lacks guidelines on the use of AI-generated political content in campaigns, leaving a regulatory vacuum that may lead to chaos in the upcoming election.
  • Since mid-June 2025, Dismislab.com has identified 70 AI-generated political campaign videos, primarily targeting Jamaat-e-Islami and other parties like BNP, NCP, and Islami Andolon.
  • These highly realistic videos simulate real interviews with 'ordinary citizens' from diverse backgrounds, presenting a fabricated image of broad public support for various parties.
  • BNP, NCP, and even AL-aligned users have started producing their own synthetic content, promoting candidates, attacking rivals, and spreading false allegations.
  • AI-generated content has misrepresented top Bangladeshi politicians, including Chief Adviser Professor Muhammad Yunus, BNP Acting Chairperson Tarique Rahman, and Awami League President Sheikh Hasina.
  • The Digital Security Act does not directly address deepfakes or AI-generated videos, leaving a gap in legal protection.
  • Bangladesh's digital literacy rate is around 10%, making it a fertile ground for disinformation.
  • Experts warn that the government or Election Commission should establish guidelines on using AI in political campaigns, and regulatory frameworks should target AI-generated misinformation, particularly in the context of the upcoming election.

Statistics:

  • 70 AI-generated political campaign videos have been identified by Dismislab.com since mid-June 2025, with a total of over 23 million views.
  • The use of AI-generated content is becoming increasingly accessible, with tools like HeyGen, DeepFaceLab, and Synthesia becoming more available for smartphone users.
  • 25 states in the United States have introduced laws to regulate the use of AI in elections, including a ban on releasing political deepfakes within a specific timeframe before voting.
  • 23 other states in the United States mandate disclosure for AI-generated elements in media content, similar to existing requirements for political advertising.
  • The European Union's AI Act categorizes AI systems into four risk levels, banning manipulative or exploitative uses and designating AI tools that can directly influence voters' choices as 'high-risk' systems.

Sources:

  • Dismislab.com
  • Dhaka Tribune
  • Qadaruddin Shishir, fact-checker
  • Fahmidul Huq, faculty member at Bard College, USA
  • Shafiqul Alam, press secretary to the chief adviser
  • European Union's AI Act
  • Texas and Minnesota statutes
  • Statutes in 23 other US states
  • Bangladesh Election Commission

Read more