Bangladesh’s growth story is not just about money. It’s about how people come together. Female education, health services, and women working were built first, then the economy grew. The Grameen model and BRAC samity were not just for charity; they were for business, and they worked because they focused on groups.
This idea is important for Bangladesh as it looks at artificial intelligence. In our experience, growth has come through social progress, not against it. We have moved away from the group-based systems that set us apart. Now, the support for youths is mostly individual: one training for one person, one loan for one person, one grant for one person. Our youths enter the AI economy alone, one freelancer dealing with one client. This does not build skills, create institutions, or help when one person fails.
We need to go back to our roots. We should fund, train, and register groups instead of individuals, starting with students. This way, young Bangladeshis can join the economy together, instead of working alone.
In South-Eastern Nigeria, a state has created such a system. It is called ÓKÓBÌ, the One Kindred One Business Initiative, launched in Imo State. The idea is simple: support a group instead of just one entrepreneur. An extended family, village, or student group comes together, chooses a business, registers it, and owns it together. To qualify, a business must be registered, owned by a group, aimed at profit, well-managed, and must create jobs.
The Imo State Office of the Chief Economic Adviser says there are 461 registered businesses and 19,676 active members in all 27 local government areas as of December 2025. These numbers come from the programme itself, so we should be careful with them. Still, the model has been recognized by the London School of Economics and partnered with Nigeria’s Bank of Industry for grants to registered groups.
One part of this model stands out because it solves a problem seen in youth grant schemes in Bangladesh. The programme explains that in individual grants, money is sometimes misused. Group ownership helps with this since everyone has a stake in the outcome. If one person manages funds alone, they can easily lose them. But if five people manage them, they watch over each other.
Rwanda has taken a similar approach with technology. A company called Digital Umuganda took the monthly community tradition of building roads and schools and focused it on digital work. They collected Kinyarwanda voice data for AI systems. Soon, Mozilla reported that Kinyarwanda became one of the fastest-growing languages on its Common Voice platform.
Bangladesh has less time to react than before with tech changes, and the pressure is growing. The Information and Communication Technology (ICT) Division set a $5 billion target for IT exports by 2025. In 2025, the sector recorded $724.6 million, with skills mismatch being one big reason. The Bangladesh Bureau of Statistics reported a 13.5 per cent graduate unemployment rate in 2025. The Business Standard calls this a long-term issue, not just a temporary one.
Despite this, research from the Oxford Internet Institute says Bangladesh is the second-largest supplier of online labor globally, with a 16 percent share, just behind India. Around 650,000 IT freelancers work in the country. But Payoneer data shows the average Bangladeshi freelancer earned between $500 and $700 a month in 2025, while a study found nearly 48 percent earned less than $208. This shows many are stuck in low-skill jobs like basic data entry. This is exactly where AI is replacing jobs quickly.
Many young people are working alone and selling their time for cheap in a market that may not want to buy it soon. Isolation is a big part of the problem, not just lack of skills. A lone freelancer can learn one tool. But a group can create a product, share tasks, serve bigger clients, and continue even if one member leaves.
This raises a big question. If working together was so natural for us, why did the digital economy grow to be so individual? Part of the answer is that freelancing platforms focus on individuals. But AI does not have to be owned by one person. We just took a path that made the individual important instead of the group. That was a mistake we can fix.
Fixing it could look like this: call it Ek Somiti, Ek Byabsha: one group, one business. Three layers, based on the Nigerian system.
A facilitator, likely the ICT Division through its Innovation, Design and Entrepreneurship Academy (iDEA) project with a bank, offers seed grants of three to five lakh taka per registered group, not per person. Universities and polytechnics act as consultants, providing training and mentoring, while the Bangladesh Association of Software and Information Services (BASIS) and freelancer groups connect them to the market. The students would own the business, registered as a cooperative or partnership, with elected leaders, audited accounts, and a profit-sharing agreement that lasts beyond graduation.
Imo State has already created a student version. Its ÓKÓBÌ Students Club has a three-module program that ends with students forming groups to pitch, test, and launch a real business. These ventures can continue after the founders graduate, helping each new group learn from the last. Our campuses could become ongoing centers for creating businesses rather than just hosting competitions that end after the awards.
What would these groups sell? Maybe AI language services, which are now in demand. Bangla itself is a great place to begin. With around 300 million speakers and many dialects, there is a lot of room for improvement in AI training data. A student group in Sylhet recording and annotating Sylheti is not just doing charity. They are creating a product that can be sold while being owned by the community. Beyond language, there are many opportunities: automation services for local businesses, Bangla tutoring, and diagnosing crop diseases by photo, among others.
We must include safeguards from the start: fair selection, professional supervision, milestone payments, independent audits, and clear rights over any data collected with informed consent from the people involved. Public funds should follow proof of customers and good governance, not just excitement.
The National AI Policy 2026 to 2030 has finished public consultation. This is the time to add a way to implement it, because a policy without action is just a piece of paper.
We need to be honest about the risks. As Policy Magazine pointed out, the national AI strategy from 2019 and 2020 had many plans that did not happen. The infrastructure is still uneven, and the current policy draft admits that less than 38 percent of rural people have internet access. This suggests starting in university towns and district polytechnics instead of rural areas.
None of this means Bangladesh must build big models or compete for tech power. It needs to find its strength again: bringing people together so technology brings shared success. The difference is that time is moving faster than before.
Drop your comment
No comments yet — be the first to drop the gist 👇