Your Old Computer Is Not Dead Yet: How Bangladesh users can save money and keep going

First time Published on The Daily Observer on June 26, 2026

For the past few years, two of the world’s biggest technology companies, Apple and Microsoft, are quietly moving to cut off support for millions of computers that still work perfectly well, especially for low to mid income people and those who really need not to run resource heavy AI tools locally. 

Microsoft has made its latest Windows 11 update incompatible with many older processors while Apple is preparing to drop all Intel-based Macs from its next major operating system, macOS 27. 

At the same time, both the companies have recently raised hardware prices significantly, Apple alone has added roughly Tk 50,000 or more to the cost of a new iMac.

For users in Bangladesh, where a mid-range laptop is worth two or three months’ salary, this creates an unavoidable problem. But it also presents a clear opportunity: a free, open-source operating system called Linux can give those same “obsolete” machines several more years of productive life- without spending a taka on new hardware.

 

Why Your Machine Is Not Actually Broken

The honest reason older computers feel slow is not the failing hardware. The truth is operating systems and helping software that have grown too resource hungry. 

Windows or macOS installed on a PC or laptop runs dozens of background programmes that consume RAM and processing power before even we open a browser. 

 

A lightweight version of Linux-based operating system, such as core Debian, Lubuntu, Bodhi Linux etc., uses less than 1 GB of memory when sitting idle. This is roughly one-third of what Windows 11 requires at rest. 

On a machine with 4 or 8 GB of RAM, a lightweight OS can give you comfortable daily use. 

 

What Linux Means for Bangladeshi Users

Linux is not a single product or an operating system. It is the core of a family of operating systems built and maintained by a global community of developers. 

It is entirely free, regularly updated, secure, and legal. Over 90 per cent of desktop malware is written to attack Windows or macOS. Switching to Linux means most of those threats simply cannot run on your machine.

For everyday tasks, writing documents, browsing the web, attending video calls, managing spreadsheets, Linux covers almost everything. 
LibreOffice, which comes pre-installed on most distributions, handles all standard Microsoft Office file formats. Google Chrome, Firefox, Zoom, and Telegram all run on Linux without any additional workarounds. 
In Bangladesh, a student, a small business owner, or a journalist with a tight budget will mostly get the benefit of a Linux-based distribution on their old machines. 

But they  will need some effort to study the official documents and tutorials online. However, installing, managing, and using Linux based distributions is not Rocket Science. 

 

How Long Can Your Device Realistically Last?

To be honest, it depends on the type of machine, but the numbers are encouraging. 

A desktop PC, the kind commonly found in Dhaka offices and residences, has no battery to degrade. Running a lightweight Linux distribution on an older desktop can realistically extend its working life by seven to ten years beyond what Windows would allow, into the 2030s, unless the power supply and motherboard remain physically sound.

Older laptops are a slightly different case. The main limiting factor is the battery. Once lithium-ion cells deteriorate beyond a certain point, the laptop effectively becomes a desktop that requires a constant power connection. 
However, these machines will outlast their chassis. A three-to-six year extension is a realistic and conservative projection for pre-2018 laptops running a lightweight Linux system.

Apple MacBooks from 2019 and 2020 require some intermediate tweakings though. 
A specialised setup through a project called T2 Linux, because Apple uses proprietary security hardware that standard Linux installers cannot read. 
The process might take an hour but, once complete, these machines, which are genuinely high-quality hardware, can run professional workloads for five to eight more years.

On AI: You Do Not Need to Run It Locally
One argument often made in favour of buying powerful new hardware is that artificial intelligence tools require significant processing capacity. But this is not true for most users.

Running a large AI model directly on your own computer demands substantial memory and a modern graphics chip. 
For the vast majority of Bangladeshi users, this is neither necessary nor cost-effective. 

Free tiers of cloud-based AI services such as Google Gemini, Meta AI, and Microsoft Copilot already handle the most common tasks like summarising documents, drafting text, translating content, answering questions. 

 

Where a paid subscription is genuinely useful, entry-level plans from the major providers currently cost between $10 and $20 per month. A small fraction of the cost of even a budget new laptop, and a fraction that covers access to far more capable models than any consumer machine could run locally.

So, keep the older machine, install a Linux-based lightweight distribution, and pay subscription if AI assistance is genuinely needed. For most of our works, free tiers are more than enough. 
That combination costs far less than a new device and delivers more AI capability than running them locally.

 

One Caution 

The only alternative that should be avoided is downloading cracked or pirated copies of Windows or macOS to bypass the new hardware requirements. 

These unofficial releases circulate widely and are a serious security liability. They receive no official security patches and are frequently bundled with malware by the people who distribute them. 

So, for your older machines, a Linux-based distribution is the safest, legal, and technically superior choice.

2,400 Crore Credentials Leaked: Why Bangladeshi users should take this seriously

First time published on The Daily Observer on June 20, 2026

Cybersecurity researchers worldwide discovered a database containing 24 billion or 2,400 crore login records that include usernames, email addresses, passwords, cookies, authentication tokens and login URLs.

As a conscious end user in Bangladesh, you cannot ignore this finding. A good number of us still use the same email address and password across multiple services, including Gmail, Facebook, online banking, mobile financial services, e-commerce platforms and government portals. 

If our credentials are found in the leaked dataset, which is not a surprise that this author himself found his email id in at least three breaches, cybercriminals can launch credential-stuffing attacks, in plain English they can use the same password to gain access to multiple accounts of the same target person.


In reality the risk is amplified by our purchasing power and digital habits. A significant number of Bangladeshi users still have to rely on older computers, unsupported and cracked versions of Windows (like 10, 8, and even still 7 and XP releases are also seen in abundance) and outdated Android smartphones due to our purchasing power and increasing price of quality electronic devices. These devices and installed operating systems often no longer receive security updates, making them easier targets for ‘infostealer malware’ a type of software developed by cybercriminals that steals saved email ids, passwords, along with browser cookies and authentication tokens to mimic the victim.

Another concern is that many users remain logged into the same Google, WhatsApp, Facebook and other accounts across multiple devices, including old smartphones, shared family computers and office PCs. If one of these devices is compromised or lost, attackers may gain access to active sessions without even needing the account password. This could allow criminals to read emails, impersonate users, steal personal data or bypass certain security checks.

 

The widespread use of shared computers in homes, cybercafés, educational institutions and small businesses also increases the risk. If users save passwords in browsers or fail to log out after using a shared device, sensitive accounts can become easy targets.

Cybersecurity experts from Cybernews believe the newly identified database is largely a compilation of credentials stolen over time through infostealer malware, that often target older devices and operating systems, rather than the result of a single company's data breach. However, that does not reduce the threat. Old credentials remain valuable because many people rarely change their passwords or continue using the same password across different services.


What can we do now? 
Let's not rush for new devices and trash the older ones. Here is a universal cyberguideline for Bangladeshi users. 

First, a user can check if her/his data is found in the leak going to this site that the researchers are offering a search option. 
Or, a user can check from Have I Been Pawned a well known site for checking stolen information.  

1. Change passwords for important accounts immediately

2. Enable multi-factor authentication (MFA) using your smart phone- Google and Microsoft Authenticator apps are the most popular and user friendly ones. 

3. Use a password manager like KeepAss or KeepAssDroid to create and maintain unique passwords safely.  

4. Regularly review the list of devices logged into their Google, Facebook, WhatsApp and other online accounts. Any unfamiliar or inactive device should be signed out immediately. Users should also install the latest security updates and replace unsupported operating systems whenever possible.

5. Instead of using cracked and outdated Windows operating systems, users can adopt Linux-based distributions like Debian or Bodhi. Linux-based distributions are optimised for older devices, and focused on user security. 
‘Linux is Hard,’ the myth has been defunct at least for a decade, to be precise from 2013-14 when it increasingly got popularity among the young generation. Moreover, some desktop environments are more user friendly and intuitive than Windows 10 or 11. 
With a little to moderate effort, a security conscious user can make her/his older hardware super fast and safe by installing a Linux distribution without any cost of buying a licence.  

6. However, using Linux based distros on smartphones are still for advanced users. But many companies in Bangladesh are offering budget smartphones with latest hardware and software for mid to low income people. By doing basic googling and consulting advanced users, one can buy a moderately secure smartphone for their daily use. 

 

How AI is reshaping global newsrooms: What Bangladesh journalists can do

First time published in The Daily Observer on June 23, 2026

Global mainstream media organizations like Associated Press, BBC, Wall Street Journal are increasingly integrating artificial intelligence into their daily workflows to increase productivity of journalists, optimize news production, automate repetitive tasks, and execute complex investigations.

However in Bangladesh, there is a significant gap in terms of institutional formalization and less focus on how we can use available free tools in newsrooms to increase our productivity.

In this article we will compare our local newsrooms with the global context. We also have some tips on how we can customize free AI tools until our newsrooms adopt their tailored AI models as per editorial policy. 

 

AI Use in Newsrooms at a glance
Rather than replacing human reporting and sub editing, some mainstream media deploy AI models to manage high-volume data analysis, generate automated copy, and refine content delivery.

Major publications like Bloomberg use proprietary large language models (LLMs) to analyze financial documents and perform sentiment analysis. Semafor uses an AI-powered feed called “Signals” to help journalists search through global news sources in multiple languages.

The Financial Times and the Wall Street Journal use AI models to predict trending topics to help journalists find gaps in their current coverage. 
The Associated Press utilizes natural language generation platforms to automatically ingest financial datasets and draft corporate earnings reports and minor-league sports articles. This enables the agency to scale coverage without draining manual reporter resources.

BBC uses internal generative AI models to reformat stories from the Local Democracy Reporting Service (LDRS) into the standard BBC house style, to generate at a glance summaries, and language translation. 

AI tools, in global media, act as smart assistants that handle, with human intervention, time-consuming tasks, allowing reporters to focus on deeper journalism, for example transcription, translation, drafting and primary editing of a news article, and multimedia creation in rush hours. 

The Digital Divide: AI Integration in Global and Bangladeshi Media
Leading international media outlets have enough budget and RnD teams to deploy custom-built and proprietary large language models in their houses. In Bangladesh, we are still in the exploratory phase of adoption.

In global media, AI adoption is guided by well-defined institutional frameworks, extensive tech budgets, and specialized software engineering teams embedded inside the newsroom. 

Conversely, in Bangladesh, AI adoption is primarily driven by individual usages under tight deadlines rather than a well structured institutional strategy. Bangladeshi journalists and writers extensively use general-purpose tools like Grammarly, Google Translate, and ChatGPT for basic proofreading, text rephrasing,creating news from press releases, and translation from Bangla to English and vice-versa.

Industry leaders at a media dialogue hosted at The Daily Star Centre said that while AI tools are highly affordable, Bangladeshi newsrooms suffer from severe ethical and operational gaps. Most local organizations operate without formal, written AI editorial policies or self-regulating mechanisms.

Editors at New Age newspaper have also noted a growing sense of limitation. They predict that news managers train staff to use generative tools like ChatGPT, however, the output frequently lacks the localized emotion, cultural nuance, and deep historical context required for impactful domestic journalism.

Cautions, Bias, and Human Intervention
The rapid adaptation of AI in journalism also brings some risks worth evaluating. 

Misinformation and Deepfakes: It is nowadays very easy to engineer highly credible fake contents, images, and videos. Without rigorous verification and fact checking, newsrooms risk unintentional risk of spreading state-sponsored propaganda, corporate manipulation, and AI-induced hallucinated facts,which can lead to a post-truth era.

Training Data Bias: Standard commercial AI models are trained on massive, and mostly western-centric datasets that carry inherent linguistic, cultural, and political biases. 
These models frequently misinterpret the historical and socio-political nuances of countries in the Global South, the Middle East, to name a few, leading to produce biased information.

These are only two of many critical problems we face in the newsroom while using AI. 
To counter these, human oversight is mandatory at every stage of the journalistic pipeline. We must prioritize accuracy over speed, quality over quantity, keeping in mind that standard tools like Google Translate, Grammarly, or ChatGPT always require strict human verification before publication.

What can we do now? 
AI, Large Language Models, and AI agents- all these are utility tools like the internet, a keyboard, a typewriter machine, or a smart assistant who can bring the best output under restricted human intervention. 

Here is an example how to chain existing AI solutions in our article writing pipeline. 
Without generating a generic story like ‘write an article on AI use in newsrooms’ and sending it to the subbing desk, we can do the following-
  
Firstly, we can use Google search or Google AI mode to collect data on how AI is used in the mainstream media newsrooms across the world and Bangladesh. This is the research and development phase to collect necessary information. 

Then we will select some five to ten original, authentic, and informative pieces.

Now, we feed NotebookLM with the sources, text, or PDFs. NotebookLM is a Google product used for creating a knowledge base from strictly defined sources so that AI makes few mistakes and does not add hallucinated data.     

Then we will filter out our necessary data (obviously cross check any critical information), and assign it the task to prepare a draft on the topic. 

Secondly, we will go to Gemini, and create an ‘article writer gem,’ a gem is your own custom built AI assistant who is specialised in a specific task like report writing or sub editing. 

Based on the information collected from NotebookLM, we will define the article writer gem how it will structure the final draft. 

After getting the final draft, we will edit and polish the article and send it to the subbing desk for further polishing. 

 

Why Bangladesh newsrooms need an OSINT habit, not just a fact-check desk

First time published on the Daily Observer on July 2, 2026     

Every major newsroom around the globe is now in the verification business to fight misinformation and reconstructing scenes to verify information.

Mainstream media like BBC built a 60-journalist unit named ‘BBC Verify’ in 2023 solely to fact-check video, counter disinformation and show audiences its evidence trail.

The New York Times is running a dedicated team named ‘Visual Investigations’ since 2017, combining digital forensics and traditional reporting. In plain English the team reconstructs crime scenes and conflict zones from cellphone footage and satellite images to verify data and information published in news.

CNN, The Washington Post and Reuters have followed with their own forensics and open-source units. This is not a buzzword, trend, or luxury, rather this is the need for an era where data passes in almost the speed of light.

Information now reaches the public before anyone in a newsroom has checked it, and audiences are increasingly unable, and do not have much time to verify, to tell real footage from fabricated footage on their own.

The verification practice is called Open-Source Intelligence, in short OSINT.

In plain English, OSINT is the practice of gathering and analysing information that is already publicly available on social media posts, satellite images, government records, leaked documents, YouTube videos, even shadows in a photograph. Then piecing them into a verified, defensible account of what actually happened.

It is not black magic, not hacking. It is not ‘espionage’. It is closer to old-fashioned shoe-leather reporting, except the “street” being walked is the internet, and the “eyewitnesses” are the millions of ordinary people filming events on their phones.

For Bangladeshi media, adopting an OSINT mindset is no longer optional and waste of time. It is becoming a rudimentary condition for publishing accurate and authentic news. This practice is equally applied for investigative reporters, as well as for us, the sub-editors who decide, often in minutes, whether a viral photo or video is safe to run under the paper’s name or portal.

A Short History: From wartime filing cabinets to Bellingcat

Interestingly, OSINT is older than the internet. During the WWII, the American Office of Strategic Services, the forerunner of CIA, ran a research branch that analyzed foreign newspapers, radio broadcasts, and even obituaries for collecting information about enemy capabilities.

The practice was intelligence gathering from publicly accessible sources. After the war, the discipline involved into a bureaucratic practice of librarians and archivists.

In 2009, young Iranians used social media to broadcast their protests to the world in near real time during Iran’s Green Revolution.

The outside observers realised that citizen-generated footages can be verified, geolocated, and used as a primary source of information.

The next major leap came from the British blogger Eliot Higgins in 2012. At that time Higgins was an unemployed office administrator. He began analysing YouTube videos of the Syrian civil war from his living room, under the pseudonym “Brown Moses.”

By cross-referencing weapons visible in the footage with satellite imagery and geography, he proved that the Syrian regime was using cluster munitions and chemical weapons, conclusions that mainstream reporters on the ground could not have reached alone.

In 2014, Higgins formalised this work into Bellingcat, a Netherlands-based collective of professional and volunteer investigators.

Bellingcat’s came in lime light within months of its founding, when it used open-source video, satellite images and social media posts to establish that a Russian Buk missile system had shot down Malaysia Airlines Flight MH17 over Ukraine that was later confirmed by an official international investigation.

Bellingcat went on to unmask the Russian agents behind the poisoning of Sergei Skripal and Alexei Navalny, using only publicly available flight records, leaked databases, and passport photos.

Bellingcat’s every step of its method was published. So, readers and rival investigators could check the workflow of the investigations. This transparency is now considered close to a scientific standard in newsrooms, and it directly inspired the large OSINT desks that BBC, the New York Times, Washington Post and others operate today.

How Mainstream Media Actually Uses OSINT

The 60-person BBC Verify brings scattered data analysis, disinformation monitoring, and fact-checking into an organized structured form. The team geolocates footages, analyses satellite imagery, checks metadata (the hidden data embedded in a photo or video file, such as when and where it was created) and cross-checks audio for signs of AI voice-cloning.

The team publishes its evidences alongside stories showing the audience how it knows something, not just what it knows.

The New York Times’ Visual Investigations team, combines traditional reporting with “open source reporting”. It geolocates a video to an exact street corner using landmarks and shadows, cross-referencing “before” and “after” satellite images to establish when a strike happened, and building minute-by-minute timelines from hundreds of scattered video clips.

In 2022, they reconstructed the killings of civilians in Bucha, Ukraine, using Google Street View imagery and metadata analysis to directly contradict Russian government claims about the timing of the killings.

The Washington Post’s Visual Forensics unit and Reuters run comparable operations, and both have collaborated with Bellingcat on cross-border investigations. CNN now runs a formal OSINT training academy for its own journalists, teaching reverse image search (tracing where else a photo has appeared online, and when), geolocation and open-source toolkit-building as baseline skills.

OSINT has two faces. It lets a desk-bound investigator do work that once required physical presence and acts as a filter against publishing fabricated content under the outlet’s name, the risk that has grown sharply with generative AI.

Why This Matters for Bangladesh

The information environment in Bangladesh has been under sustained strain since the July 2024 uprising, and it intensified through the campaign for the 13^th parliamentary election held on February 12, 2026.

Rumor Scanner alone verified 310 separate misinformation cases in July 2025, rising to 320 in August and 329 in September, with well over half of all cases in each month tied to political content. Across the country’s fact-checking organisations combined, more than 5,700 fact-check reports were published in 2025 alone, roughly 30 percent more than the year before. Two out of every three of those fact-checks involved political content, and roughly one in ten used artificial intelligence to fabricate the deceptive material in the first place.

The period immediately around the February 2026 election was especially intense. On the single day before voting, three fact-checking organisations, Rumor Scanner, Dismislab and FactWatch, jointly documented 49 unique pieces of election disinformation in under 20 hours, including fabricated “news photocards” mimicking the visual style of established outlets falsely claiming candidates had withdrawn or that results were already decided.

For Bangladeshi editorial desks, the lesson is direct, the country’s evolving political and social climate means that speed, virality and partisanship are now baked into the information environment, and a newsroom that cannot independently verify what is landing in its inbox and its social feeds is at permanent risk of amplifying a fabrication, deliberately planted or not.

The Problem for Reporters and Sub Editors: Old Footage, Deepfakes and Rumour

In Bangladesh, fact checkers documented three recurring and common patterns that every reporter and sub-editor should be able to recognise on sight, even without sophisticated tools. These are the most rudimentary basics of utilizing OSINT in journalism.

Zombie Contents: In plain English, these are old footages recycled as breaking news. This is the single most common tactic documented around the 2026 election.

For example, a viral clip showed an election-day clash in Gopalganj. Fact checkers traced it to an August 2024 rally in Narayanganj.

The defence here is reverse image search. Taking a screenshot from the video or the photo itself and running it through a search engine’s image tools (Google Images, Bing, or specialist tools like TinEye) to see where and when it first appeared online. If the same clip surfaces in a report from a year earlier, it is not a breaking news.

Deepfakes:  Deepfakes are AI generated image, video, or audio contents.  Analysis of the July to September 2025 period found 71 AI-generated posts in circulation, the large majority of them videos, 57, and the rest were 14 images.

A fabricated image can be referred that shows students demanding the return of former prime minister Sheikh Hasina circulated in August 2025. It was entirely synthetic. The basics to identify a deepfake content is to look for unnatural blinking, mismatched lighting or shadows, blurred edges where a face meets hair or background, and audio that doesn’t quite sync with lip movement.

However, AI tools are improving quickly and becoming more natural looking. So, cross-checking against a second, independent source matters more than any single visual clue.

Fabricated “photocards” and impersonated outlets: Photocards are now a trend in social media to attract more user interactions to posts. During the national electionsin 2026, the creation of fake graphics designed to look exactly like a real outlet’s branded news card, like mimicking mainstream Bangladeshi media, was seen to fabricate instant credibility.  Herer the defence is procedural, not technical. Any screenshot of “news” from another outlet should be verified directly on that outlet’s own website or verified social account before republishing or citing it.

Fact-Checkers Exist, So Why Do Newsrooms Still Need OSINT?

Bangladesh already has an established and functioning fact-checking ecosystem. Rumor Scanner, Dismislab, FactWatch, BoomBD, and AFP Fact Check Bangladesh to name a few. So, it is logical to ask why every reporter and sub-editor needs to acquire OSINT skills when specialist teams already do this job.

The answer is a division-of-labour, not a duplication of efforts.

Fact-checkers, by design, work after a claim has already gone viral. Their reports are fast and rigorous and are inherently reactive. Rumor Scanner’s fastest turnaround on election night was still measured in hours, not the seconds it takes for a sub-editor to decide whether to publish a breaking photograph.

A newsroom’s own desk is the only line of defence before a news is printed or goes online. The moment a wire photo, a viral clip or a tip-off first lands in the newsroom, someone on deadline has to decide whether it will be published or not. A fact-checker debunking an outlet’s story the next morning is reputational damage that basic verification, done in-house, could have prevented in the first place.

Bangladesh’s fact-checking organisations are committed but they are a handful of small teams checking a national information ecosystem that produced over 5,700 flagged claims in a single year.

It is unrealistic that they check everything crossing every a newsroom’s desk in real time, nor should reporters expect them to. Basic-to-moderate OSINT literacy like reverse image search, checking metadata, confirming a claim against a primary or official source, knowing how to geolocate a photo is what allows a reporter or sub-editor to filter the truth.

In May 2026, panels of senior editors at Bangladesh’s own Journalism Conference in May 2026 identified that modern investigative journalism needs financial forensics, data analysis, cybersecurity awareness, digital verification, and source protection that many newsrooms still inadequately support.

A newsroom in Bangladesh is right now navigating a volatile political transition and a surge of AI-generated fake contents. The basic layer of in-house scepticism and OSINT skills is no longer a specialist skill, rather, a must for both reporters and sub-editors.