The speakers revealed that scaling AI technologies across the organization has been challenging for them on many fronts
Financial services have become increasingly digitized introducing new opportunities and challenges. With more digital financial services, customers expect more innovative experiences.
To deliver innovative financial services, companies are looking to enhance their data platform to leverage public and private clouds to provide a secure, controlled, and customizable experience for storing and processing confidential data.
Financial services companies also want to leverage artificial intelligence and machine learning (AI/ML) technologies to make more data-driven decisions to improve customer experience and mitigate risks. And they need to be able to do this in a safe and secure manner with all of their data.
ET Edge Insights, in association with Cloudera and Hoonartek, held a Data Leaders roundtable in Mumbai on the topic of “Trusted Data is the Basis for Trusted AI.”
The roundtable witnessed participation from Data Science and Analytics leaders from the financial services space who shared their views on the challenges around connecting data with business goals, and why trustable foundational data and models are critical for building a competitive advantage for a future driven by Generative AI.
The speakers revealed that scaling AI technologies across the organization has been challenging for them on many fronts. Some of the factors that hindered their progress were the absence of a coherent AI strategy, an outdated and rigid technology core, scattered data assets, and obsolete operating models that prevented business and technology teams from working together.
On the emergence of Generative AI, the speakers were of the view that while the new technology has benefits, it also poses many challenges and risks, such as data privacy, contextual limitations, ethical concerns, and model governance.
The speakers said that they must deal with various issues, such as:
Data protection: The models use a lot of data, some of which may be private or confidential. The models need to be safeguarded and managed well to prevent data breaches or abuses.
Situational constraints: The models may not always generate reliable or suitable outputs, especially in complex or unclear scenarios. The models need to be checked and supervised to avoid mistakes or misunderstandings.
Moral issues: The models may create content that is unfair, harmful, or deceptive, such as false information, manipulated images, or fake identities. The models need to be consistent with the values and principles of the organization to avoid ethical breaches or reputational harm.
Model management: The models are harder to explain, audit, or regulate than traditional AI models, because of their complexity and variability. The models need to be transparent and accountable to avoid non-compliance or liability.
Financial services companies need to adopt a responsible by design approach that follows five principles: purposeful, fair, secure, explainable, and governable. This way, they can use Generative AI to enhance their productivity, efficiency, and customer experience.
Piyush Agarwal, SE Leader, Cloudera India explained how the company is addressing these challenges by offering a secure and scalable hybrid data platform that allows enterprises to train and deploy their own Large Language Models (LLMs) using their proprietary or regulated data sources within a secure framework. “We also provide tools and best practices for managing the lifecycle and quality of LLMs, ensuring that they are responsible and reliable,” he said.
Agarwal added that Cloudera’s Open Data Lakehouse is the foundation for Generative AI in the enterprise, as it enables access to trusted data that is secured, governed, and ready for AI. Cloudera also offers Cloudera Machine Learning, which includes Applied ML Prototypes (AMPs) that help accelerate AI experimentation and development with LLMs. Cloudera’s customers include OCBC Bank, Experian, Deutsche Telekom, and others who use generative AI to enhance their products and services.
Amit Raj, Chief Strategy Officer at Hoonartek stated that AI and automation are pivotal in reshaping the industry, but their true potential lies in harnessing the power of data. “Having a robust data strategy plays a crucial role in enhancing the impact and ensuring long-term sustainability,” he said.
