Federation of banks survey shows growing adoption to enhance efficiency in more than half of institutions
04/19/2024
/i.s3.glbimg.com/v1/AUTH_37554604729d4b2f9f3eb9ad8a691345/internal_photos/bs/2024/g/Q/lgjw5xQSaEwmBPRLZzRg/sergio-biagini-deloitte.jpg)
Sergio Biagini — Foto: Divulgação
The Febraban 2024 Banking Technology Survey, conducted by consulting firm Deloitte at the request of the Brazilian Federation of Banks, shows that 96% of banks have artificial intelligence technologies, and 54% say they already use generative AI (GenIA). According to the study, GenIA is highly adaptable and customizable to the specific business needs of each institution.
While traditional AI only copies, imitates, or reproduces something that already exists or has been done, GenIA can create new and original content based on what it has learned.
“This technology has the potential to drive innovation and productivity, as well as promote efficiency. The research indicated that for banks that have already implemented AI, the efficiency of banking processes has increased at an average rate of 11%,” said Sergio Biagini, Deloitte’s lead partner for the financial services industry.
Among the AI applications used by banks are facial biometrics (75%), chatbots (71%), robotic process automation (67%), GenIA (54%), and cognitive intelligence (25%).
“With the need to seek an increasingly relevant position towards differentiation, banks can use the Open Finance movement to expand data and knowledge about their customers. With the use of technologies such as AI and GenIA, information can be explored more efficiently and accurately to personalize customer relationships,” said Mr. Biagini.
The study points out that, like any other type of technology, GenIA has some limitations, which can be considered challenging, such as hallucination (creating responses with full conviction but not based on its training data); biases (inherited from its training data); lack of logical reasoning (based on statistical characteristics, which does not allow human logical reasoning); and contextual limitation (composed of a finite combination of words and contexts, which limits the generated results).
“Banks are leading adopters of generative AI, especially in customer service processes and developer support, which has allowed an improvement in the efficiency of banking processes,” the research points out. The efficiency increase brought by AI is estimated at 11%. In the back office, this index is even higher, at 17%.
Concerning expectations of the adoption of the Open Finance initiative, banks continue to work to offer new products and services related to the ecosystem and aim to achieve a 6% to 20% adherence rate from their active base by the end of 2024.
*Por Álvaro Campos — São Paulo
Source: Valor International