IMF and OECD Warn AI Could Widen the Divide Between Rich and Poor Economies

IMF and OECD Warn AI Could Widen the Divide Between Rich and Poor Economies

Artificial intelligence could help the global economy grow. Countries that don’t have strong infrastructure or enough workers with tech skills might not get much of the benefit. Officials from the International Monetary Fund (IMF) and the Organisation for Economic Co-operation and Development (OECD) said this on October 22.

It’s not about whether companies use AI. It’s about whether whole economies can use it and at a large scale. Things like electricity, good internet, workers who know how to use the tools and the ability to fit AI into regular business work will help decide which countries gain the most.

The IMF says that if AI is used well it could add up to 0.5 percentage points to economic growth each year over the next few years.. This is only a possible outcome, not something that’s already happening. IMF experts also say that clear proof of productivity gains from AI is still missing.

The OECD thinks AI could boost productivity growth by 0.4% to 1.2% per year in its member countries. Its chief economist, Stefano Scarpetta pointed out that many people use AI tools for tasks but businesses aren’t using them deeply in their main operations yet.

Infrastructure could decide who wins

AI doesn’t just run on software. Businesses need power, internet, computing power, data and workers who can actually use the tools. These things are harder to get in places where electricity's unstable or internet is too expensive or not evenly spread.

For countries this could mean a two-speed transition. Companies in countries might be better able to use AI in research, finance, logistics, customer service and manufacturing. In prepared markets businesses might pay more to adopt AI or not have the systems needed to turn AI access into real output.

The IMF has shared warnings before. A 2025 working paper looked at AI’s impact and found that advanced economies could see growth effects more than twice as big as low-income economies in the model it used. The paper said a country’s exposure to AI how ready it is to use it and access to tech and data all matter.

This is a model-based idea, not a guarantee that every developing country will fall behind. Good policies and investments can help spread the benefits widely.

Another IMF study from July 2026 found that in developing economies AI value was mostly concentrated in a group of professionals. Based on the data and measures used that suggests just having access to AI tools isn’t enough. The tools need to be used across jobs, more businesses and more parts of the economy.

Productivity gains are still a guess

The growth numbers come with questions. AI can help people write documents, sort information write code and automate routine tasks.. How much the economy benefits depends on how companies change their work to fit these new tools.

Putting in a tool doesn’t mean productivity jumps away. Companies may need to change their processes fix data problems, train workers and set up rules before they see results. In some places the benefits might take time to appear.

The OECD’s warning about use in core business functions is important. Just because people use chatbots or other AI tools for use doesn’t mean companies are using AI deeply enough to boost the whole economy.

This difference matters for leaders and investors who are judging the effects of the current AI boom. These forecasts describe what could happen if AI is adopted well and broadly. They shouldn’t be taken as proof that the gains are already happening.

Young workers may face a shift

The discussion also covers jobs. AI can. Help with tasks that used to help junior workers learn on the job—like writing routine reports summarizing documents answering common questions or doing basic research.

IMF official Era Dabla-Norris said younger and entry-level workers could be at risk if these tasks are taken over by AI. If companies hire junior workers or expect new hires to already have skills that used to be learned on the job the first steps into a career could get harder.

We need to be careful how we read the evidence. The OECD’s 2026 report on people and jobs says there’s no clear proof yet that AI is causing worse outcomes for young workers. Other things like slowdowns hiring trends in certain industries and more also affect early-career jobs.

So exposure to AI doesn’t mean a job is gone. Some jobs may change some tasks may. New duties may come up. The effects will depend on the job the company and how fast organizations adopt AI.

What governments and businesses can do

These warnings point to actions. Governments can fix electricity and internet make digital tools easier to get and support training that helps workers use AI well. Schools might need to focus on digital skills thinking critically solving problems and abilities that work with automated tools.

Businesses also have a role. Training staff to work with AI can help companies improve output without cutting jobs. Employers can look at how entry-level jobsre set up so automation doesn’t remove every chance for junior workers to build experience and judgment.

For developing countries the challenge is bigger than building a tech industry. AI could help in farming, public services, healthcare, education and small businesses.. How far it helps depends on local infrastructure, cost, language support and real training.

The next test is adoption

The IMF and OECD warnings point to a big question for the global economy: will AI lead to broad productivity growth that many people share or will it mostly help countries and companies that are already better prepared?

The possible gains are big. They depend on many things. Governments need to build the basics for AI use. At the time businesses and schools must help workers adapt to changing tasks and career paths.

The important thing to watch isn’t just how fast AI tools spread. It’s whether their use leads to lasting productivity improvements, wider economic inclusion and real paths into jobs, for workers.