Ziauddin Liton

"I wake up and see a video. It's me—yet I'm saying things I never said. Has reality slipped beyond my control?"

Once seen as a beacon of liberation, technology now raises questions about control, truth, and identity. The rise of artificial intelligence (AI)—especially in generating hyper-realistic videos—is blurring the line between reality and illusion. If fake becomes indistinguishable from real, where does that leave truth?

Today’s AI is not just analyzing data—it’s crafting deep fake videos, cloning voices, generating virtual presenters, and even replicating a person’s face and tone with unsettling precision. These creations can deceive not only the general public but also seasoned professionals.

Yet, the story is not all grim. In Bangladesh, platforms like 10 Minute School and Shikho are deploying virtual teachers to transform digital education. Health startups are using AI videos to spread vital awareness in rural areas. Organizations like YPSA are producing descriptive, subtitle-supported videos for persons with disabilities—marking a positive stride toward digital inclusion.

However, the threat is growing just as rapidly.

AI-generated videos are becoming so authentic that discerning truth from fiction is a growing challenge. As misinformation spreads, trust erodes—in media, in democracy, and in justice. In 2023, several political deep fake videos circulated on platforms like Facebook and YouTube, misleading hundreds of thousands of people.

AI is also redefining professions. News anchors, teachers, actors, and even brand ambassadors are being replaced by digital avatars. This shift could displace millions of workers. Moreover, as a few tech companies or powerful states gain control over this technology, economic and power inequality may deepen.

The emotional impact is no less alarming. Perfect AI-generated beauty, ideal voices, and controlled expressions may lure people away from real human connection. We may witness a rise in identity crises, loneliness, and depression. A society built on artificial relationships and synthetic feelings risks losing the essence of what makes us human.

Politically, the stakes are even higher. Ahead of elections, during protests, or in volatile moments, AI-generated speeches or provocative videos can be weaponized—spreading rumors, inciting unrest, and threatening democratic stability.

The Election Working Group (EWG) warned in its 2024 report:

“AI-generated misinformation is likely to influence public sentiment unless actively countered.”

Clearly, regulations must catch up. Globally, countries are waking up. The European Union is introducing the AI Act, while the U.S. Department of Defense is working on Sema For—a project to detect and label AI-generated content. Tech giants like OpenAI, Google, and Meta have committed to mandatory watermarking.

Bangladesh, however, lacks a dedicated legal framework for AI content. Neither the ICT Act (2006) nor the Cyber Security Act (2023) includes specific provisions for AI-generated visuals or voices. It's time to draft a strong AI Content Regulation Policy, including:

Mandatory watermarking of all AI-generated videos, Legal penalties for using someone’s face or voice without consent,

Compulsory media literacy education starting from the school level.

The solution lies not in fear but in foresight. We must foster ethical AI practices, enhance content verification technology, harmonize global tech regulations, and prioritize emotional and moral education.

AI carries immense promise—but also profound peril. If the boundary between truth and fiction vanishes, if humanity dissolves into synthetic perfection, then we risk becoming shadows of the machines we’ve built.

AI should not be our rival. It must become our ally. Let reality not surrender to illusion. Let truth not be buried under digital lies. Let this be the spirit that guides our shared future.

The author is educator, journalist, and columnist.He can be reached at email litonsir12@gmail.com