Beyond Business as Usual: The Value of an AI Strategy

Ciaran Fennessy (Citi) discusses AI’s escalating influence in the funds industry and how organisations must shift their perspective from the standard ‘Business as Usual’ approach to a well-constructed, articulated and focused AI strategy becomes imperative.

The vast majority of companies, across all industries, are giving significant consideration to AI. No matter the industry, AI is becoming more and more prevalent. Some of this can be traced back to the explosive growth in interest following the November launch of ChatGPT 3.5. For many, there had already been significant awareness and investment in the technology and what it can achieve to drive business strategies.

Given AI’s growing pervasive presence across all these sectors, including the funds Industry, organisations need to ensure their AI strategy goes beyond the conventional ‘Business as Usual’ approach. The emphasis needs to pivot to the value delivered by an AI strategy and what should organisations be cognizant of.

Prior to diving into an AI strategy, let us first determine what constituents a good strategy. According to Richard Rumelt, in ‘Good Strategy/Bad Strategy: The difference and why it matters’1, a good strategy boasts a trifecta of key elements ”a diagnosis, a guiding policy, and coherent actions”.

Applying Rumelt’s approach to an AI strategy, the initial step involves identifying the diagnosis. Specifically, what is the obstacle you are looking for AI to address? Where precisely is the challenge at hand? Clearly defining and articulating this obstacle ensures that efforts are directed appropriately, paving the way for deriving value from an AI strategy.

The ‘guiding policy’ pertains to the methodology adopted to address and surmount the challenges outlined in the diagnosis. In the context of an AI strategy, this guiding policy is anchored in data. Every organisation houses data across diverse systems. Yet, one could argue, there is a pressing need to shift the organisational perspective on data.

Bill Schmarzo notes, in ‘The Economics of Data, Analytics and Digital Transformation’2, “The value of data is determined by how you use it to create new source of value”. Schmarzo recommends that organizations change their approach from being ‘data-driven’ to being ‘value-driven’. Therefore, having a ‘value-driven’ mindset on data needs to be a fundamental tenant within the guiding policy.

The final pillar are the coherent actions. These are the actions that will operationalise the guiding policy, ensuring that the value from your AI strategy is realised. Obviously, these actions very organisation specific, however, the following points merit consideration;

- How capable is your organisation at using data to drive business decisions?

- What is required to achieve a ‘value-driven’ mindset?

- What needs to be complete to enhance this?

- What insights can you get from your data?

Ethics, not usually a consideration for technology strategies, is a critical component for an AI strategy and one organisations need to be acutely cognisant of.  Whilst organisations within the Irish funds industry may not be developing real-time facial recognition systems, it is critical that the impact of AI initiatives be considered and the approach reflected in the AI Strategy. 

Indeed, in 2022, the Bank of England defined it’s model risk management principles for banks.  Within this document, the Bank of England articulates their expectations for firms to meet five model risk management principles, with the principles reflecting the full end to end project lifecycle for machine models.

- Model definition – Firms have an established definition of a model that sets the scope for MRM, a model inventory, and a risk-based tiering approach to categorise models to help identify and manage model risk.

- Risk Governance – Firms have strong governance oversight with a board that promotes an MRM culture from the top through setting clear model risk appetite.

- Lifecycle Management – Firms have a robust model development process with standards for model design and implementation, model selection, and model performance measurement.

- Effective Challenge – Firms have a validation process that provides ongoing, independent, and effective challenge to model development and use. 

- Model Risk Mitigants – Firms have established policies and procedures for the use of model risk mitigants when models are under-performing, and have procedures for the independent review of post-model adjustments.

A discussion on AI Strategy is incomplete without focusing on Generative AI.  This technology promises to have a transformational impact across an organisations AI strategy, amplifying the value derived from it.  According to McKinsey research, estimates are that generative AI can add $2.6 trillion to $4.4Trillion annually, with 75% of this delivered across use cases in

-  Customer operations

-  Marketing and Sales

-  Software Engineering

-  R&D

While media coverage suggests that emerging technologies, such as Generative AI, might threaten job security, recent studies suggest otherwise.  MIT Sloan reported that “employees derive individual value from AI when using the technology improves their sense of competency, autonomy, and relatedness. Likewise, organisations are far more likely to obtain value from AI when their workers do.”

Indeed, within the field of software engineering, research from GitHub on the use of GitHub Co-Pilot (a generative AI-powered code assistant that helps developers write code more efficiently), found that 90% of developers using GitHub Copilot report completing tasks faster, with 75% of developers who use GitHub Copilot feel more fulfilled and able to focus on more satisfying work.

As AI’s escalating influence in the Funds Industry continues to proliferate, organisations must shift their perspective from the standard ‘Business as Usual’ approach.  To support this, the value that can be driven from a well-constructed, articulated and focused AI strategy becomes imperative.  Recognising the ethical consideration of the AI solutions developed and the necessity for the governance need to support the regulatory climate becomes critical too.  As note by Chris Young, in his article ‘Building a Winning AI Strategy for Your Business’, the “leaders who embrace AI now … and envision how it can solve hard problems are going to run companies that thrive in an AI world.”

References

1 Rumelt, R., 2012. Good Strategy, Bad Strategy: The Difference and why it matters. s.l.:Profile Books Limited.

2 Schmarzo, B., 2020. The Economics of Data, Analytics, and Digital Transformation: The theorems, laws, and empowerments to guide your organization's digital transformation. s.l.:Packt Publishing.