The Bank of Ghana is expanding its use of artificial intelligence, Big Data, and machine learning to improve economic forecasting and identify financial-sector risks earlier, as policymakers seek faster and more precise intelligence for decision-making.
First Deputy Governor Dr Zakari Mumuni said the central bank had deployed AI and Big Data technologies to develop its in-house electronic inflation nowcasting methodology, known as e-Inflation, while machine-learning models are being used alongside conventional econometric models to forecast GDP and conduct text-mining analytics.
The move is changing how the Bank gathers and interprets economic information, particularly as the volume and speed of available data increase.
“The greatest challenge facing policymakers today is no longer a shortage of data. It is turning an abundance of data into timely, reliable and actionable intelligence,” Mumuni said at the 4th Annual Statistics and Data Science Conference of the Ghana Statistical Association in Tamale.
He said technology was particularly valuable when it reduced the time between an economic event occurring and policymakers becoming aware of it.
The Bank is also applying more granular data in financial supervision. According to Mumuni, supervisors previously relied heavily on static monthly spreadsheets that required manual reconciliation. Increasingly, data can now be validated as it arrives, allowing potential risks to be identified earlier.
“That is not a cosmetic upgrade,” he said. “It can mean catching a problem while it is still small rather than discovering it after it has grown.”
The growing use of alternative data could also broaden the information available to policymakers. Mumuni pointed to digital payment activity, tax records and satellite imagery as potential sources of faster signals on consumption, business activity and agricultural conditions.
The use of these technologies could strengthen the Bank’s monetary policy process, as its inflation-targeting framework requires the Monetary Policy Committee to assess where the economy is, where it is heading, what could change its trajectory and what policy response is appropriate.
That process involves analysing prices, output, credit, exchange rates, fiscal conditions and financial markets, while considering how the indicators interact. The Bank uses econometric techniques and its Quarterly Projection Model within a Forecast and Policy Analysis System to assess trends, risks and possible policy outcomes.
However, the First Deputy Governor cautioned that greater computing power does not eliminate the need for reliable underlying data or human judgement.
“AI can process enormous volumes of data, but it cannot turn bad data into good data,” he said, stressing the importance of sampling, measurement, classification, validation, metadata and revisions.
The Bank is therefore also focused on improving the quality and relevance of the data feeding its models. Mumuni said changing consumption patterns and the emergence of new economic sectors meant statistical measures had to be periodically updated, citing the rebasing of GDP and the Consumer Price Index as important examples.
He also noted that the growing use of payment, tax and telecommunications data creates opportunities but brings confidentiality and data-governance responsibilities.
The Bank’s approach reflects a broader effort to make policy more forward-looking. Governor Dr Johnson Pandit Asiama had committed the institution in February 2025 to a more proactive and precise approach to managing inflation through advanced data analytics and artificial intelligence.
Mumuni said the Bank had since continued to strengthen its capabilities in artificial intelligence, data analytics and virtual assets.
He argued that technology should complement rather than replace professional judgement, with statisticians, economists, data scientists, technologists and policymakers working more closely together.
“Technology can strengthen our intelligence, but it does not remove the need for human judgment,” he said.
The objective, he said, is to ensure that research moves beyond publication into practical policy, with better data supporting faster decisions and more effective economic outcomes.
