Where AI can help Sri Lanka earn more from its exports
“Innovation is the specific tool of entrepreneurs, the means by which they exploit change as an opportunity for a different business or a different service.”
— Peter F. Drucker, Innovation and Entrepreneurship (1985).
By Prof. Asoka S. Seneviratne
Introduction
In his recent article on Professor Prema-chandra Athukorala’s research into Asian production networks, Dr. W. A. Wijewardena makes a suggestion worth pursuing. Sri Lanka could use artificial intelligence and advanced design tools to strengthen its role in garment design. His example raises a wider question for policymakers. At which stages of a value chain can AI help Sri Lankan firms solve a problem for overseas buyers and retain more of the value they create? Read Wijewardena’s article in the Daily FT.
A value chain runs from (i) product development and (ii) the sourcing of inputs through production, (iii) delivery and (iv) after-sales service. Sri Lanka’s opportunity is to identify the stages where its firms already have expertise and where AI could help them offer something more valuable to customers.
Apparel can show how this works
Consider a Sri Lankan apparel manufacturer asked to develop a new garment for an overseas buyer. Its design team could use AI-assisted tools to explore options and prepare digital prototypes before making physical samples. If that reduces the number of sampling rounds, the buyer can reach a decision sooner, and the manufacturer can spend less time and material on revisions.
A larger opportunity arises if the buyer gives the Sri Lankan firm responsibility for more of the product development. The firm could then earn income from design and development alongside manufacturing. That outcome depends on its ability to deliver designs the buyer wants, meet technical specifications and negotiate payment for the additional work. Those are commercial tests a pilot project can help answer.
The distinction matters for export policy. Saving time and material can improve the competitiveness of an existing export. Earning payment for additional work can increase the value of that export. In either case, the benefit to Sri Lanka depends on how much value remains in the country after paying for imported inputs, software, and other services.
Look for opportunities beyond apparel
The same approach could be applied to tea and processed food. A tea exporter could test whether AI-assisted quality checks help it identify inconsistent batches earlier and meet a buyer’s specifications more reliably. A processed-food exporter could use demand forecasts to plan smaller, more frequent shipments and reduce spoilage. The commercial benefit would show up in fewer rejected consignments, lower losses or stronger repeat orders.
In logistics, a Sri Lankan service provider could test whether predicting shipment delays gives overseas clients enough notice to change delivery plans. In professional services, an engineering practice could use AI to help review technical documents while keeping qualified engineers responsible for the final work. If clients value the faster turnaround, the practice may be able to serve more overseas customers or take on more complex assignments.
The starting point in each sector is a problem the customer already recognises. That gives firms a way to judge whether AI is worth the investment.
Give firms a way to test the commercial case
The Export Development Board (EDB) could work with industry associations to bring exporters and overseas buyers together around a few clearly defined problems. A first round of pilots could focus on apparel sampling, food quality or shipment reliability. Each pilot should involve a firm willing to test the application and, wherever possible, a buyer willing to assess the result. Universities and training institutions could help firms develop the technical skills needed to run and evaluate the trials.
The Board of Investment (BOI) could help connect local suppliers with investors and international firms able to provide technology, expertise and access to customers. The National Business Facilitation Centre (NBFC) could help resolve administrative obstacles that surface during the pilots. For smaller exporters, shared testing facilities and training could make it possible to evaluate a tool before committing to a costly subscription or equipment purchase.
Firms should also agree in advance on how they will handle confidential designs and customer data, and who will verify AI-assisted work before it reaches a buyer. These decisions are part of delivering a service customers can trust.
Measure what Sri Lanka gains
Each pilot should record a starting point and measure what changes. In apparel, that could mean the time and cost of preparing samples, buyer acceptance rates, and revenue from design work. In food exports, it could mean fewer rejections and repeat orders. Where a firm earns additional export revenue, the evaluation should also estimate how much of that revenue stays in Sri Lanka after the cost of imported technology and services.
The broader lesson from Wijewardena’s example is that AI can help Sri Lankan firms take responsibility for more of the work that overseas buyers value. If a garment producer can earn from product development as well as manufacturing, or a logistics company can sell a more reliable service, Sri Lanka has a practical route to higher-value exports. A few carefully measured trials would give policymakers and firms evidence about where to invest next.
How reshipment can generate more income locally
Sri Lanka could begin with a pilot at the Port of Colombo to test how reshipment can generate more income locally. Where customs rules and buyer requirements permit, goods passing through Colombo could be inspected, sorted, labelled for different markets, repacked or combined into buyer-specific consignments before being shipped onward. AI could help logistics firms anticipate connecting vessels’ arrival times, flag discrepancies in shipment documents and identify consignments that could be combined to reduce costs. These services would give overseas buyers a commercial reason to route more cargo through Colombo while generating income for Sri Lankan firms. The pilot should track turnaround times, shipment errors and the additional income retained in Sri Lanka after the cost of providing the services. The economic test is how much more Sri Lanka earns from each shipment it handles.
Summary
Sri Lanka can use artificial intelligence to strengthen its export industries by solving specific problems for overseas buyers. In apparel, AI-assisted design could help firms develop and revise samples faster. In tea and processed food, AI could help exporters detect quality defects earlier and deliver more consistent products. In logistics, it could help firms anticipate delays and give customers more reliable delivery estimates. The economic opportunity is to earn more from exports by taking on activities that buyers value, including design, quality assurance and specialised services. Policy should begin with practical trials and judge them by measurable results such as shorter lead times, fewer rejected shipments and higher export earnings. How reshipment can generate more local income must also be noted.
Conclusion
Sri Lanka’s starting point should be a question that exporters can put directly to their overseas buyers. What problem could we solve for you that would make our product or service more valuable? The answer may be faster sample development, more consistent quality or a more reliable delivery date. Exporters, technology firms and universities can work together to test where AI makes a measurable difference, while the Export Development Board can help smaller firms learn from trials that succeed. The ultimate test is whether(i) Sri Lankan enterprises can win more business, (ii) take on higher-value work and (iii)n retain a greater share of the FX income they help generate.

(The author served as the Special Adviser to the Office of the President of Namibia from 2006 to 2012 and was a Senior Consultant with the UNDP for 20 years. He was a Senior Economist with the Central Bank of Sri Lanka [1972-1993]. He can be reached via asoka.seneviratne@gmail.com)