This series explores the coming of GenAI and how it has been framed in a governance context thus far. It also touches on the differences in definitions of AI Governance and how Unique defines AI Governance for Financial Services, namely, our AI principles and their operationalization. It encompasses the implementation through processes, procedures, policies, and regulations to ensure that FinanceGPT aligns with our values.
This is the first part of our AI Governance Series which will focus on:
Finally, we discuss the importance of AI Governance for banks and insurances and how it is necessary to business value and efficiency because it mitigates data privacy, intellectual property, and ethical risks associated with AI, and it enhances organizational access to and adoption of AI systems, thus driving innovation and process efficiencies.
When ChatGPT was released in 2022, Generative Artificial Intelligence (GenAI) rapidly became a hot topic. Despite AI having been around for several decades, this was the first conscious interaction many people had with AI. Quickly this led to a broad public debate around the concerns surrounding the technology. Both government and private companies rapidly began upscaling their efforts to understand and manage this technology. Especially in the financial services industry (FSI), many players soon started to design and implement first use cases with GenAI. In this context, Unique FinanceGPT is a European market leader providing a tailored solution for the financial industry that aims to increase productivity by automating manual workload through AI and ChatGPT solutions.
With the introduction of a new technology like ChatGPT and other GenAI solutions, new risks and responsibilities arise, especially for highly regulated industries like banking or insurance. One of the most important topics is around AI Governance to ensure correct, unbiased and factually valid outputs of GenAI solutions.
This topic arose long before the rise of ChatGPT and has been gaining momentum, with a dramatic increase in AI Governance frameworks having been published since 2016. These range from international instruments like the OECD AI Principles and the UNESCO Recommendations on AI Ethics; to frameworks and laws by governments like the EU AI Act and the AI Bill of Rights by the US government; to frameworks laid out by private companies or organisations like the Institute of Electrical and Electronics Engineers (IEEE)’s Ethically Aligned Design, Microsoft’s Responsible AI Principles, IBM’s Principles for Trust and Transparency, and Google’s AI Principles.
There are currently many ongoing discussions and implementations of AI governance frameworks, including initiatives by private sector coalitions, international organisations, individual governments and companies. There has also been more work in implementing AI governance in and for specific industries and companies including in the Financial Services Industry (FSI) and here at Unique.
Despite the plethora of definitions, AI governance is often referred to as the frameworks, policies, and practices that guide the development, deployment, and management of artificial intelligence (AI) technologies. Its goal is to ensure that AI systems are developed and used in ways that are ethical, transparent, accountable, and aligned with societal values and laws. It includes topics like ethical guidelines, regulatory frameworks, technical standards, industry standards, accountability mechanisms, risk management and more.
AI governance within a single company can vary greatly in its scope, in what it encompasses and who is responsible for it. In summary, AI Governance for Unique means "our AI principles and their operationalization. It encompasses the implementation through processes, procedures, policies, and regulations to ensure that FinanceGPT aligns with our values”.
With the rise of new technologies like GenAI, the question arises why we need a certain governance around it. Like all governance, AI governance is essential to provide alignment between organisational strategy and objectives. Further, it assists in ensuring the management of risks associated with AI.
AI and GenAI specifically in our case, has massive potential to enhance customer experiences, improve operational efficiency, lend a competitive advantage to the developer and promote data-driven decision-making. However, despite all these great benefits, they do not come without risks. In a survey undertaken by McKinsey in 2024, organisations indicated that they had experienced negative consequences to the organisation due to risks including inaccuracy (23%), cybersecurity (16%), explainability (12%), regulatory compliance (10%), personal/individual privacy (9%) and more. They also listed the following as the relevant risks associated with GenAI: Inaccuracy (63%), cybersecurity (51%), compliance (43%) and explainability (40%).
From a business standpoint, AI Governance is necessary as it is critical to business value and efficiency, it mitigates data privacy, intellectual property, and ethical risks associated with AI, and it enhances organizational access to and adoption of AI systems, thus driving innovation and process efficiencies.
Currently, many FSI companies are using AI without having a defined AI governance since GenAI use case deployment is still very new to most banks and insurance. This can result in a misalignment between the company’s goals and the way GenAI is being used by employees or deployed to customers. It is crucial to ensure that GenAI-driven technologies are aligned with legal requirements, regulation and ethical considerations that are important for a highly regulated industry like banking and insurance.
We have seen a rapid increase in AI regulation across the world in the past few years and this doesn’t seem to be stopping any time soon. Therefore, it is essential for any banks and insurances using AI to not only ensure alignment with its own goals but also with the external constraints that are given by local and international regulations. AI governance includes putting in place the necessary compliance mechanisms to ensure that the company does not have to pay fines, go to court or suffer from damages resulting from negative press.
Therefore, it can be said that with AI Governance banks and insurances can balance value creation and risk mitigation related to AI whilst ensuring that AI is used in a manner that aligns with its values, goals and strategies.
With Unique FinanceGPT, banks and other financial services get a build-in AI Governance framework which is built both on strong technical and legal foundations. Stay tuned for the second edition of this white paper Series.