Financial institutions collect vast amounts of structured and unstructured data on a daily basis, and AI is presenting more efficient ways to utilize that data — opening analysis and innovation potential for the industry. 

Roughly 80% of the data Citigroup sees is unstructured — from emails, contracts and regulatory documents — and that data can be extracted using machine learning and AI, Prag Sharma, global head of AI at Citi, said today during the webinar “Global ideas for better banking,” presented by Bank Automation News. “That 80% of data is now in play like it never was before,” he said.

With access to broader use of unstructured data, use cases and innovation can take off, Jennifer Warren, head of markets digital strategy at Barclays, said during the webinar.  

“The ability to leverage both structured and unstructured data to extract sentiment analysis and market color, and be able to interrogate that data is incredibly useful when it comes to productivity and better service,” she said. 

To bring AI innovation to life, financial institutions must bring together business architecture, technology architecture and that data architecture, EJ Achtner, global head of AI for commercial banking at HSBC, said during the webinar. 

For example, financial institutions can use AI to summarize content, compare documents, extract key information and translate documents, Sharma said. 

“Much more of your digital data is suddenly in play to be analyzed with machines to a level that was never possible before,” he said. “The use cases from that [AI] are only limited by your imagination.”