Pro-government news outlets and influencers in Bangladesh are promoting AI-generated disinformation created with cheap tools ahead of the January 2024 elections
Benjamin Parkin / Financial Times : X: @1br0wn , @areeqchowdhury , @tobywalsh , and @valayism X: Ian Brown / @1br0wn : “It's easy for a politician to say, ‘This is deepfake’, or 'This is #AI-generated” @sabhanazdiya said. “The challenge for the global south...is going to be how the idea of AI-generated content is being weaponised to erode what people believe to be true” https://www.ft.com/... Areeq Chowdhury / @areeqchowdhury : “A deepfake video, which has since been removed, showed an opposition leader equivocating over support for Gazans, a potentially ruinous position in the Muslim-majority country with strong public sympathy for Palestinians.” https://www.ft.com/... Toby Walsh / @tobywalsh : So predictable ... deepfakes mislead voters in Bangladesh elections. Remember that there are critical elections next year in India, Europe, US, UK .... and little is being done to prevent more deepfake disruption ... https://www.ft.com/... Valay Vaidya / @valayism : Interesting read mentioning how the deepfakes can mislead voters and how it can fuel fault lines in the society. Police & Law Enforcement Agencies have greater challenges ahead with the increasing usage and normalisation of these technologies! https://www.ft.com/...
Context & Ripple Effects
Bangladesh joins an emerging election-integrity pattern in which synthetic media is used not only to fabricate claims but to make authentic reporting easier to dismiss. The tactic echoes deepfake videos circulated before Slovakia's parliamentary election, showing how cheaply produced content can be adapted to local political fault lines.
The policy backdrop remains uneven: early deepfake rules in China contrasted with limited regulation elsewhere, leaving platforms, political actors, and voters to navigate attribution disputes without consistent safeguards.
First-order effects
- Pro-government outlets and influencers can rapidly distribute AI-made material that damages opposition figures or exploits sensitive issues ahead of the vote.
- Voters face a more difficult credibility test: the availability of synthetic content also gives politicians a ready basis to deny inconvenient authentic material as fake.
Second-order effects
- Opposition campaigns, journalists, and platforms must spend more resources verifying and rebutting contested media, while removals can arrive after a claim has already spread.
- The Bangladesh case reinforces pressure for election-specific platform responses, even as broader deepfake regulation remains contested on free-speech grounds.
Third-order effects
- If low-cost production and strategic distribution continue to combine, election disinformation may shift from isolated viral fakes toward a sustained contest over whether any political evidence is trustworthy.
- The durable governance challenge is dual-use: measures that improve provenance and accountability must avoid becoming tools for suppressing legitimate political speech.
The trend: Cheap generative tools are lowering the cost of political influence operations while weakening the shared evidentiary baseline on which election debate depends.