A third-year female student at Dhaka University switched off her phone in the middle of the night.
Hours earlier, she had commented on a political post on Facebook. Soon after, her inbox filled with abuse and threats – messages saying “we will deal with you”, “we know your address”, and even threats of rape.
The following day, she received calls from unknown numbers repeating the same language. Since then, she has struggled to attend classes regularly. Fear, anxiety and uncertainty have disrupted her daily life.
Her experience is no longer an isolated case.
Across Bangladesh, online political discourse – marked by abusive language and hate speech – is becoming increasingly hostile, with consequences spilling into the real world.
Saiful Alam Chowdhury, associate professor of Mass Communication and Journalism at Dhaka University, told TIMES of Bangladesh that the problem runs deeper than the platforms themselves.
“The issue lies more in political culture than in the medium – the tendency to demean others and use provocative language,” he said.
He added that the impact of such attacks is now directly affecting people’s lives beyond the digital space.
Supporters of the country’s major political forces are increasingly locked in online confrontation. As the political rivalry between BNP and Jamaat-e-Islami intensifies, their followers are also clashing online.
There are also growing allegations of manipulated and AI-generated “deepfake” images being circulated to target individuals.
In some cases, online posts are leading to legal action, arrests and even physical violence.
Recently, an image depicting Prime Minister Tarique Rahman and his daughter Zaima Rahman spread on Facebook, triggering clashes between student wings – Chhatra Dal and Chhatra Shibir – at Dhaka University, in Cumilla, and at a college in Chattogram.
Against this backdrop, concern is also growing internationally. On 16 March, Amnesty International warned that unless Meta takes timely and effective action to tackle harmful content on Facebook, Bangladesh could face more serious human rights violations.
There is evidence that social media has been used to mobilise attacks on religious communities and sites in the country.
On 11 April, a group attacked a Sufi shrine in Daulatpur upazila of Kushtia, killing a spiritual leader, Abdur Rahman, also known as Shamim. Prior to the attack, Facebook posts had circulated accusing him of blasphemy.
On 18 December last year, violent attacks were carried out on the offices of two leading newspapers, The Daily Star and Prothom Alo. In the months leading up to the assaults, threats against the organisations had been widely shared on social media.
Posts openly labelled them “agents of India” and “anti-state forces”, while calling for attacks and arson. The direct link between online incitement and real-world violence was evident.
Many of the abusive and provocative posts that trigger outrage appear to spread in a coordinated manner. In numerous cases, content is shared within 30 seconds to a minute of being posted, before rapidly being amplified by thousands of accounts, suggesting organised dissemination.
There are also allegations of so-called “bot networks” aligned with political groups. Critics say these networks target opponents with waves of abuse, while also spreading defamatory content against rival factions.
Chhatra Dal President Rakibul Islam Rakib told TIMES that even ordinary incidents are being distorted and presented out of context.
“Organised groups are deliberately spreading such distortions and propaganda to conduct what amounts to a ‘media trial’,” he said.
“Without understanding the full context, people are constructing their own narratives from a single moment or image,” he added.
When asked about similar allegations against supporters of his own organisation, the student leader declined to comment.
Recently, in a television talk show, allegations were raised that certain groups and pages were spreading misinformation in a coordinated manner and using “bot accounts”, with direct and indirect involvement of leaders from Chhatra Shibir.
At the programme, the claims were rejected by Shibir leader Shadik Kayem, who said there was no evidence linking his organisation to rumour-mongering or the operation of fake pages.
He instead accused supporters of Chhatra Dal of running propaganda campaigns and character attacks through various Facebook pages.
The vice-president of Dhaka University Central Students’ Union said a “zero tolerance” policy should be adopted against bot accounts and organised online networks, adding that the issue should be addressed through legal measures.
Language growing more aggressive
Amid an increasingly hostile online environment, everyday interactions are becoming more aggressive, according to Monirul Islam, a motorbike ride-sharing driver in Dhaka. He says abusive and violent language is now more common, even in ordinary conversations.
“What we see on Facebook, people now repeat on the streets,” he said. “The language of the virtual world is shaping real-life behaviour.”
Gazipur-based college teacher Ayesha Amin Khuku said political debate has also shifted in tone.
“Earlier, political arguments were about reasoning. Now they turn straight to insults. Students are picking up the same behaviour,” she said.
Former Senior Vice-President of the Bangladesh Association of Software and Information Services Mostafizur Rahaman Sohel said growing political polarisation is pushing people into an “us versus them” mindset.
“In that space, attacking the other side starts to feel normal,” he said.
Prime Minister’s Press Secretary Abu Abdullah M Saleh also expressed concern over the tone and civility of language used on social media.
“The words now being used online in the name of protest do not align with the society’s established human and cultural standards,” he told TIMES.
“The kind of abuse and derogatory language being used in the name of free speech does not reflect a healthy society,” he added.
Attacks on women take multiple forms
A significant share of political posts on social media now contain direct insults or personal attacks. For women, the abuse is often more severe – targeting their character, families, and even their personal safety.
A private university student in Dhaka, Toma Jaman, said the level of abuse intensifies when she writes about politics.
“When I post on political issues, the harassment increases several times over. Not only me – my family is also targeted,” she said.
Many people now feel compelled to think twice before posting, weighing the possible backlash. As a result, the space for free expression is gradually shrinking.
Abusive posts or screenshots shared online are also affecting real-life relationships. Friends, colleagues and family members often become entangled, with long-term consequences for mental health.
Data from an initiative called Cyber Support for Women and Children Platform shows that 41.3 per cent of cyber violence victims are students, 20.7 per cent are homemakers, and 10.5 per cent include sales workers, NGO employees and businesspeople.
What is the way forward?
Experts are calling for media literacy education to be introduced from the school level to help address the challenges of online hostility.
“People need to learn how to verify information, engage with differing views through reason, and build a healthy language and culture both online and offline,” said Associate Professor Saiful Alam Chowdhury.
Secretary of the Information and Communication Technology Division Kazi Anwar Hossain said while the issue does not fall entirely under a single ministry, steps are being considered.
“We are taking into account the spread of coordinated, aggressive posts and online conflicts,” he said.
He added that a briefing would be held soon to address relevant aspects, including those under the Posts and Telecommunications Division.
Efforts are also under way to engage with social media platforms, he said.
Ahmed Yaseer Abrar, research officer at fact-checking organisation Dismislab, said misinformation must be tackled without undermining freedom of expression.
“Whenever false information is identified, it should be verified,” he said.
“If the misinformation relates to the government or administration, the authorities should respond with accurate information to inform the public,” he added.
