From seemingly authentic videos of homes collapsing into rivers to fabricated footage of tragic events, social media is now a breeding ground for misinformation powered by generative AI. Experts are raising concerns over the unchecked spread of such content, warning that if not addressed promptly, the consequences could be disastrous.

Videos created with AI tools are increasingly realistic—so much so that they are often indistinguishable from real events. One recent example involves digitally resurrecting deceased individuals in fabricated video messages, complete with life-like voice and visuals, misleading viewers into believing false narratives.

While some dismiss these clips as obviously fake, social media comment sections often tell a different story. Thanks to convincing titles and advanced editing, many users accept them as fact, fueling the viral spread of false information.

“Misinformation spreads six times faster than the truth,” says Dr. Aftab Hossain, a mass communication specialist. He adds that AI-generated content also poses a serious threat to user privacy. "Once personal data like national IDs become public, it creates serious vulnerabilities," he warns.

Read more: Over 11,000 YouTube Channels Shut Down Over Misinformation

With Bangladesh preparing for its 13th national election, concerns are mounting about AI’s potential misuse in political campaigns. Chief Election Commissioner A.M.M. Nasir Uddin recently described AI manipulation as a "major challenge" ahead of the polls.

Crime analysts echo these concerns. Dr. Touhidul Haque warns that opportunistic groups could exploit AI to distort facts, disrupt public perception, and manipulate election outcomes. “Law enforcement must use advanced tech to track and expose these actors to set a public example,” he suggests.

Read more: TikTok Removes Over 11 Million Videos from Bangladesh

Tech experts argue that regulatory frameworks are urgently needed. Cybersecurity analyst Tanvir Hasan Zoha points out that AI's growing influence is already evident. “Once you share any information online, there’s currently no safeguard in place to control how that data is used in the future,” he explains. He urges policymakers to fast-track AI regulations currently in draft under the Cyber Security Act.

Read more: I Swear—Internet Shutdowns Are a Thing of the Past: Tayeb

While AI offers undeniable benefits, specialists emphasize the urgent need for responsible research and policy to prevent its misuse. The time to act is now.