GenAI has become this year’s buzzword in the industry, with banks, finance firms and other companies rushing to embrace the relatively young technology.
While 2023 was the year the world discovered GenAI, 2024 may go down as the year many financial services organisations and other entities started to put it to use.
However, more and more firms are starting to realise that returns on GenAI investments are taking much longer than previously anticipated, with data being singled out as the biggest obstacle.
Some companies even abandon projects after the proof-of-concept stage due to inadequate data guardrails, a lack of real-time access to fresh, multi-source data, and escalating costs, underscoring the critical role of data in the success of GenAI initiatives.
In fact, while most organizations are moving beyond GenAI exploration to pilots, but not to production.
Some 43% of businesses are involved in department-level pilots, only 13% are conducting organization-wide data-driven GenAI pilots, according to new research shared with QA Financial.
Moreover, only a tiny fraction, 2%, of respondents have advanced to production deployment, even lower than recently reported by industry analysts, a survey by K2view, a data integration company specialised in real-time data integration, governance, and delivery.
Data
Among the top challenges enterprises are facing as they work toward generative AI (GenAI) implementation is the use and implementation of data, particularly in the areas of data accessibility and latency, data privacy, and security, the K2view team found.
They said that enterprise data is at the crux of the biggest challenges for AI deployment, with close to half of respondents citing data security and privacy concerns, and 33% cite enterprise data readiness as roadblocks to deployment.
The difficulty lies in the fragmented nature of enterprise data, which is often spread across multiple systems and analytical data stores, making it hard to integrate, govern, and make accessible in near-real time, under stringent guardrails, to GenAI applications.
“Data is critical to realizing AI’s true potential, but organisations are struggling to close the GenAI data gap and move their GenAI projects to production,” explained Ronen Schwartz, the chief executive officer of K2view, which was founded by Achi Rotem and Rafi Cohen in 2019.
“Organisations are struggling to close the GenAI data gap and move their GenAI projects to production.”
– Ronen Schwartz
Palo Alto, California-based Schwartz stressed his firm “conducted this survey to get to the heart of the issues that business and IT leaders are running into when implementing GenAI, and that prevent them from delivering sustained business value from GenAI production deployments.”
The K2view team further found that Retrieval Augmented Generation (RAG) is widely adopted as a framework for grounding LLMs with internal data, but enterprise application data is largely underutilized in initial RAG implementations.
By utilising advanced technologies like RAG, generative AI can analyse a variety of sources, offering an up-to-date perspective on trends, pattern and sentiments.
A majority of respondents, six out of ten, are already piloting RAG, while just under a quarter are in the planning stages, and 12% have moved to full production.
The survey found that “a unique set of challenges specific to enterprise application data” hinders its use in making GenAI and RAG initiatives successful, even when the data is stored in data lakes and data warehouses.
The K2view study was carried out in July of this year by collecting volunteer responses from 300 senior professionals in the U.S. and UK who are directly involved in the planning, building, or delivery of GenAI applications. Participants were drawn from B2C industries, including banking, insurance, and telecommunications, among others.
NEXT WEEK
QA FINANCIAL FORUM LONDON: RECAP
Last month, on September 11, QA Financial held the London conference of the QA Financial Forum, a global series of conference and networking meetings for software risk managers.
The agenda was designed to meet the needs of software testers working for banks and other financial firms working in regulated, complex markets.
Please check our special post-conference flipbook by clicking here.
READ MORE
- Taking place tomorrow: the QA Financial Forum Chicago 2025
- HDFC Bank turns to Katalon and QualityKiosk for QA upgrade
- Perforce’s Clinton Sprauve on AI testing for charts and graphics
- Functional and regression testing crucial for banks, says insider
- Shastic teams up with MeridianLink to target banks with AI automation
Become a QA Financial subscriber – for FREE
* Receive our weekly newsletter * Priority invitations to our Forum events
