Watchdogs flag AI misinformation in Bangladesh polls
Hundreds of undisclosed AI-made political videos are circulating online ahead of Feb. 12 polls, alarming watchdogs and election officials
DHAKA, Bangladesh (MNTV) — Artificial intelligence–generated political campaign videos are spreading widely across social media platforms in the run-up to Bangladesh’s parliamentary election on Feb. 12, heightening concerns over misinformation, voter manipulation and the integrity of the electoral process.
An investigation by the independent fact-checking organization Dismislab found that hundreds of AI-created videos portraying ordinary citizens endorsing political parties have been circulated online, often without clear disclosure that the content was artificially generated.
According to the findings, the videos typically depict elderly people, rural women or everyday workers speaking directly to the camera and praising specific parties or electoral symbols.
In several widely shared clips, individuals appear to promise economic security, welfare benefits or justice under particular political banners — narratives designed to resonate strongly with undecided or economically vulnerable voters.
Closer analysis revealed technical patterns commonly associated with AI-generated content, including repetitive scene resets and abrupt camera angle changes at regular intervals. Despite these indicators, the videos were presented as authentic recordings of real voters and distributed as campaign material without transparent labelling.
Between Jan. 1 and 15, Dismislab documented more than 800 AI-generated political videos posted across Facebook, YouTube and TikTok. On Facebook alone, 576 such videos were traced to 21 separate pages, with around 60% carrying no indication that they had been produced using artificial intelligence.
Even where disclosure existed, it was inconsistently applied. Dismislab found that Facebook’s AI information label appeared on some versions of the mobile app but was absent when the same videos were viewed through Facebook Lite or desktop browsers. Similar gaps were identified on other platforms, where the overwhelming majority of analysed videos lacked any AI-related warning.
The content promoted multiple political actors using fabricated characters such as police officers, journalists, minority community members or children to reinforce credibility.
The Election Commission of Bangladesh has publicly flagged concerns about AI-driven misinformation on more than one occasion in recent weeks, warning that emerging technologies could be misused to influence voters ahead of polling day.
Researchers involved in the investigation cautioned that low digital literacy levels increase the risk that voters may mistake AI-generated videos for genuine testimonials. They also noted that during election periods, confirmation bias plays a powerful role, with users more likely to trust and share content that aligns with their political preferences while ignoring verification.
As generative AI tools become cheaper and easier to deploy, analysts say Bangladesh’s election highlights a broader global challenge: how democratic systems can respond to rapidly evolving technologies that blur the line between authentic political speech and synthetic persuasion.
We must take charge of our narrative. Research indicates that 92% of their Muslim coverage is negative. Help us change that.
Have Your Say!
Have you been affected by the issues raised in this story? Please share your experiences by emailing haveyoursay@soundvision.com.
Please include a contact number if you are willing to speak to a Muslim Network TV’s representative. You can also get in touch in the following ways:
Tweet: @MuslimNetworkTV
