{"id":47173,"date":"2026-03-10T19:20:41","date_gmt":"2026-03-10T16:20:41","guid":{"rendered":"https:\/\/mk.gen.tr\/banks-adopt-ai-fast-but-lack-strategy-and-governance\/"},"modified":"2026-03-10T19:20:41","modified_gmt":"2026-03-10T16:20:41","slug":"banks-adopt-ai-fast-but-lack-strategy-and-governance","status":"publish","type":"post","link":"https:\/\/mk.gen.tr\/en\/banks-adopt-ai-fast-but-lack-strategy-and-governance\/","title":{"rendered":"Banks adopt AI fast but lack strategy and governance"},"content":{"rendered":"

A new survey of financial institutions suggests that although banking<\/a> is rapidly adopting artificial intelligence<\/a>, it still lacks the strategy, infrastructure and governance needed to scale the technology responsibly.<\/p>\n

A recent whitepaper released by Wolters Kluwer<\/a><\/strong> found that about 61% of financial institutions have either deployed AI or machine learning tools in production or are actively piloting them. However, only 12.2% said their organizations have a \u201cwell-defined and resourced\u201d AI strategy, highlighting a gap between adoption and long-term planning.<\/p>\n

The findings are based on a survey of 148 financial institutions and indicate the sector is at a \u201ccritical inflection point\u201d as firms move from experimentation toward broader deployment of AI systems.<\/p>\n

Operational efficiency remains the primary driver of AI investment, with 46.6% of respondents saying that their main goal is reducing costs or improving internal processes. Just 10.1%, however, cited competitive advantage, and 6.8% prioritized improvements to customer experience. <\/p>\n

Zorina Alliata, director of AI enablement at Wolters Kluwer, told HousingWire<\/strong> that the first few years of AI adoption were characterized by experimentation and proof of concept. Now, Alliata says, the \u201crubber is hitting the road,\u201d and companies aren\u2019t sure how to transition from concepts to clear strategy.<\/p>\n

As a result, just 9.5% of respondents said their data infrastructure is \u201cvery prepared\u201d to support AI initiatives, while nearly half described their systems as only somewhat prepared. <\/p>\n

Data quality was cited as the biggest challenge, followed by integrating AI with legacy systems and navigating regulatory requirements.<\/p>\n

\u201cUnfortunately, the data infrastructure gap is quite real. I\u2019ve been doing machine learning and AI for like 10 years now, and data is the biggest issue every time. If you don\u2019t have good data, then you don\u2019t have any wisdom,\u201d Alliata said.<\/p>\n

Banks are also largely deploying AI in defensive use cases tied to risk management. The most common applications include risk management, fraud detection, customer service chatbots and compliance<\/a> monitoring. Credit underwriting lags, in part because of regulatory concerns around model validation and fair lending.<\/p>\n

Regulatory uncertainty continues to weigh on adoption. Only about 26.4% of institutions said they are confident they can align AI initiatives with regulatory requirements, while the majority reported only partial or uncertain confidence. <\/p>\n

Tailoring AI to your business<\/h2>\n

Alliata says that AI adoption and initiatives will only be as good as the expert knowledge building them. \u201cYou need experts and expertise to make the AI smart. If you don\u2019t have that domain knowledge fed into the AI, your AI will never be good enough\u2026your AI needs to understand your specific company.\u201d<\/p>\n

Explainability and transparency of AI models ranked as the most significant regulatory concern, followed by risks related to bias, discrimination, data privacy and fair lending<\/a> compliance.<\/p>\n

Governance frameworks are also still developing across the industry. The report found that 46.6% of institutions have centralized data governance structures in place, while about a quarter said they lack such frameworks or are unsure whether they exist. <\/p>\n

Similarly, just over a third of respondents reported having formal policies governing ethical AI use.<\/p>\n

Financial institutions said clearer regulatory guidance would be the most helpful factor in advancing AI strategies, followed by technical training and industry benchmarks.<\/p>\n

\u201cOne common misconception that I think we\u2019re seeing in the survey is that data has to be perfect before you can do AI. Granted, if your data is bad, your AI robots will not work well. However, if you wait for all of your data to be perfect, well, I\u2019ve never seen perfect data anywhere,\u201d Alliata said. \u201cSo what we\u2019re telling people is to pick one thing, just one use case, and fix the data for that.\u201d <\/p>\n

The whitepaper concluded that the next two years \u201cwill separate AI leaders from laggards.\u201d Alliata imagines that risk management will see the largest growth in the next 24 months due to the need to stay ahead of bad actors<\/a> in the space. <\/p>\n

Despite the challenges, adoption is expected to continue growing. About 54.8% of institutions said they plan to expand their use of AI within the next two years, though nearly 39% remain uncertain about their expansion timelines.<\/p>","protected":false},"excerpt":{"rendered":"

A new survey of financial institutions suggests that although banking is rapidly adopting artificial intelligence, it still lacks the strategy, infrastructure and governance needed to scale the technology responsibly. A recent whitepaper released by Wolters Kluwer found that about 61% of financial institutions have either deployed AI or machine learning tools in production or are…<\/p>\n","protected":false},"author":0,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"inline_featured_image":false},"categories":[1],"tags":[],"_links":{"self":[{"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/posts\/47173"}],"collection":[{"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/comments?post=47173"}],"version-history":[{"count":0,"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/posts\/47173\/revisions"}],"wp:attachment":[{"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/media?parent=47173"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/categories?post=47173"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mk.gen.tr\/en\/wp-json\/wp\/v2\/tags?post=47173"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}