Vigilance on AI Abuses Needs to Be Further Fortified

Artificial intelligence is rapidly becoming part of everyday life in Bangladesh, transforming education, business, journalism, health care, banking and public services. AI tools can improve productivity and expand access to information, but their rapid adoption also creates new opportunities for fraud, misinformation, privacy violations, discrimination and other forms of abuse.

For Bangladesh, the challenge is particularly significant. A large and increasingly connected population is gaining access to generative AI tools while regulatory, institutional and public awareness mechanisms are still developing. The country therefore needs stronger vigilance to ensure that technological progress does not come at the expense of individual rights, public trust and social stability.

The risks are not hypothetical. Generative AI can produce convincing text, images, audio and video within seconds. Deepfakes can make people appear to say or do things they never did. Fraudsters can use AI-generated voices or messages to impersonate relatives, officials, businesses or financial institutions. False information can be produced and distributed on social media at a scale that would have been difficult to achieve before the arrival of generative technologies.

Bangladesh's large social-media population makes this problem particularly important.

A New Generation of Digital Risks

The misuse of AI can take many forms.

One concern is misinformation. AI-generated material can make false claims appear credible, particularly when users do not have the tools or knowledge needed to verify what they see online. During periods of political tension, natural disasters or public emergencies, fabricated photographs and videos could spread rapidly and complicate efforts to provide accurate information.

Financial fraud presents another danger. Criminals can combine AI-generated content with social engineering to create highly convincing scams. A person may receive a message apparently from a family member asking for money, or a voice call that appears to come from a colleague or company representative.

The growth of digital financial services in Bangladesh makes vigilance particularly important. As more people use mobile financial services and online banking, protecting digital identities and financial accounts becomes increasingly essential.

Privacy Cannot Be an Afterthought

AI systems also raise questions about personal data.

People routinely upload photographs, documents, conversations and other information to digital platforms without fully understanding how their data may be processed. Businesses and institutions may similarly use AI systems to analyze customer or employee information.

Bangladesh needs clear standards governing the collection, storage, processing and sharing of personal data in AI-related applications.

The country's Personal Data Protection Ordinance, 2025, represented an important step toward establishing a legal framework for personal information. The government has described the framework as intended to strengthen data protection and provide greater rights and safeguards concerning personal data.

However, laws alone cannot guarantee protection. Effective enforcement, independent oversight, technical expertise and public awareness are equally important. AI and the Workplace

AI's impact on employment deserves careful attention as well.

Bangladeshi companies are increasingly using automation and AI for customer service, content production, data analysis, software development and administrative tasks. These technologies can increase efficiency and create new opportunities, but they may also alter existing jobs.

The appropriate response is not simply to resist automation. Workers need opportunities to acquire new skills and adapt to changing labor-market requirements.

Universities, vocational institutions and employers should expand AI literacy and practical training. Young people entering the workforce should understand not only how to use AI tools but also how to evaluate their accuracy, protect confidential information and recognize manipulated content.

Protecting Education

Education is another area requiring vigilance.

AI can help students explain difficult concepts, practice languages, conduct research and develop ideas. Teachers can use it to prepare materials and personalize learning.

At the same time, unrestricted use can encourage plagiarism, fabricated references and dependence on automated answers. Students may submit AI-generated assignments without understanding the material.

Bangladesh's education system therefore needs clear policies rather than blanket reactions. Schools and universities should teach students how to use AI responsibly, disclose its use when appropriate and verify information independently.

The objective should be AI literacy, not simply AI prohibition.

Journalism Faces a Difficult Test

The media sector has an especially important responsibility.

AI can assist journalists with transcription, translation, research and data analysis. But it can also generate fabricated quotations, photographs and videos that may deceive audiences.

News organizations in Bangladesh should establish internal verification procedures for AI-generated or AI-assisted material. Journalists should confirm important claims through reliable sources and avoid publishing synthetic media merely because it appears convincing. Readers also need greater media literacy. A sophisticated-looking image is no longer proof that an event actually happened.

In an environment where misinformation can spread faster than corrections, verification must become a routine part of digital citizenship.
Stronger Institutions Are Essential

Bangladesh's approach to AI governance should involve more than policing individual abuses after they occur.

Government agencies, technology companies, universities, civil-society organizations and law-enforcement bodies need mechanisms for sharing information about emerging risks.

Kum Kum Roy studied Business Administration in Metropolitan University, Sylhet.