Future of Global Fintech Second Edition 2025

Among the risks assessed, data breach and privacy ranked highest, with 84% classifying it as a moderate to very high risk. AI-generated deepfakes were rated as a moderate to very high risk by 83% of fintechs, while cyberattacks were similarly rated as a moderate to very high risk by 79% of fintechs. Exacerbating biases and discrimination (74%), systematic risk in financial systems (71%) and market-wide concentration (71%) represented the bottom three risks, with respondents ranking these as moderate to very high. This trend was generally consistent across regions, with data breaches and privacy issues and AI- generated deep fakes representing the top two risks in the moderate to very high risk level. Data breach and privacy risk were dominant in Europe, with 90% of fintechs ranking this as moderate to very high. In MENA, cyberattacks were the primary concern, with 86% of fintechs ranking them as a moderate to very high risk. Notably, the AI-generated deepfakes and exacerbating biases and discrimination risks were particularly prominent in SSA, with 96% of fintechs ranking these as moderate to very high. SSA also dominated in market-wide concentration and systematic risk in financial systems, with 89% of respondents ranking both as moderate to very high. Fintechs operating in AEs reported data breach and privacy as the highest risk, with 87% ranking this as moderate to very high. AI-generated deepfakes were reported as the highest risk by fintechs operating in EMDEs, with 84% ranking this as moderate to very high. By vertical, insurtech firms had the highest percentage of negative perceptions (with consistently high percentages in the moderate to very high categories) across all risks analysed – from 85% in systematic risk in financial systems to 92% in data breach and privacy. Barriers to AI implementation Assessing obstacles to AI implementation stemming from internal and external factors can offer insight into what’s still needed to create a robust enabling environment. To better understand such challenges, the survey asked fintechs about perceptions of a set of barriers to AI implementation. The cost of implementing and maintaining AI systems remained a major challenge for most fintechs, with 87% of firms identifying it as a hurdle (41% considering it a significant hurdle and 46% viewing it as a slight hurdle). Notably, 47% of fintechs in EMDEs and 52% of digital payments firms deemed implementation costs a significant hurdle. Data access and quality, encompassing issues such as data sparsity and lack of variety, posed a challenge for 69% of fintechs, with 48% seeing it as a slight hurdle and 21% considering it a significant one. This concern was particularly pronounced in SSA, where 40% of fintechs cited data as a significant hurdle hindering AI implementation. Overall, 31% of fintechs did not perceive data access and quality as a hurdle, highlighting that this was not a universal challenge. Fintechs did not see other internal factors, such as unclear value propositions or a company’s culture and governance, as major hurdles. Fintechs in EMDEs expressed more confidence in these areas, with 48% seeing no hurdle in defining AI’s value proposition and 52% reporting no challenges related to culture and governance. By vertical, 62% of insurtechs were confident in AI’s value proposition, while 71% of digital bank and savings firms reported no culture and governance challenges. In regard to external factors, regulatory uncertainty surrounding AI was a major hurdle for fintechs. A substantial 77% of fintechs perceived it as a challenge, with 32% considering it a significant hurdle and 45% viewing it as a slight one. Only 23% of fintechs did not perceive regulatory uncertainty as a hurdle. Regionally, regulatory uncertainty was a particularly significant concern for fintechs in APAC (35%), LAC (36%), the US and Canada (33%) and SSA (35%). While access to skilled talent is a key driver of fintech growth (as seen in Chapter 3), it presents a challenge for AI implementation, potentially limiting firms’ ability to develop AI-driven solutions. Although 51% viewed lack of talent as only a slight hurdle, it was a significant barrier for 49% of digital lending fintechs. Other external factors, such as trust and user adoption of AI and market uncertainty, remained a hurdle for 74% and 60% of fintechs, respectively. Notable regional variations emerged. Europe had the most concern over trust and user adoption of AI, with 37% viewing it as a significant hurdle, while 33% in SSA cited market uncertainty as a significant obstacle hindering AI implementation.87% of firms identified the cost of implementing and maintaining AI as a hurdle. The Future of Global Fintech: From Rapid Expansion to Sustainable Growth 44