Why banks in Ghana make little use of AI for climate action decisions

A new study conducted at the Brandenburg University of Technology Cottbus-Senftenberg (BTU) in collaboration with the Durban University of Technology in South Africa has investigated the reasons behind this. The study was published in the prestigious journal Discover Sustainability by Springer Nature.

Enormous sums are now being channelled into climate protection worldwide. In 2024, the figure stood at around two trillion US dollars. The problem is that almost none of this money reaches sub-Saharan Africa, even though the region is particularly hard hit by climate change and is in urgent need of funding. One reason for this is that banks and investors often lack reliable data and effective tools to assess which climate projects are genuinely worthwhile and secure.

Prof. Herwig Winkler from the chair of Production Administration and Management, and co-author of the study, sees a solution: “This is precisely where AI could help, for example by automatically assessing risks or analysing sustainability data. So the possibilities are there. It’s just that hardly anyone has made use of them so far.”

What is needed for AI tools to actually be deployed 

To find out why AI adoption is so low, the researchers surveyed 317 employees at Ghanaian banks, insurance companies and investment firms. 

The result can be summarised simply: “The key factor is whether the respondents see a clear benefit in AI and whether they trust it,” adds the researcher. “Whether a piece of software is easy to use, on the other hand, plays a much smaller role.”

Consequently, this means that if a company is well-positioned – that is, has clear rules, possesses modern technology and employs trained staff – easy-to-use AI software has a significantly more positive impact on acceptance than in less well-positioned firms. In terms of the perceived benefits alone, however, this made hardly any difference.

The biggest obstacles

When asked about the biggest obstacles, respondents cited three main factors: a lack of technical infrastructure, insufficient in-house expertise and poor data quality. Uncertainty regarding legal regulations and high costs followed only as secondary concerns. Concerns about the fairness or transparency of AI decisions, on the other hand, were mentioned less frequently. This is probably because the foundations for using AI are simply still lacking in many organisations.

What the researchers recommend

The authors derive practical recommendations from the findings: Ghana’s financial regulator should establish clear rules for the use of AI in the climate sector. Banks and insurance companies should provide targeted training for Staff Members. And policymakers should invest in better climate data in collaboration with private partners.

The question of why people adopt new technologies – or choose not to – arises just as much on the factory floor as it does in a bank’s credit committee in Accra. The level of participation shows that Cottbus’s research on this topic is in demand internationally – far beyond traditional industry.

The study was carried out in collaboration with the Durban University of Technology in South Africa. Dr Emmanuel Ahatsi is currently working as a visiting researcher on a scholarship from the German Academic Exchange Service (DAAD) at the Production Administration chair in the Department of Industrial Management at BTU. 

The study is freely accessible (Open Access).

Source: Ahatsi, E., Winkler, H., Olanrewaju, O.: “Artificial intelligence-driven sustainable climate finance decision-making in Ghana’s financial sector.” Discover Sustainability 7, 1333 (2026). DOI: 10.1007/s43621-026-04407-y