In the media

Five ways banks can develop an AI-ready operating model

By Alexandra Valentine, Aaron Barolia, Dan Waites

Finance Derivative

21 November 2023

The AI revolution is upon us, with far-reaching implications. Financial services will be one of the most disrupted sectors, and banks are already seeing improvements to their top and bottom lines through AI. A recent Harvard Business School study concluded that a cohort using AI was significantly more productive and delivered higher-quality results, completing 12% more tasks, 25% faster.

But fully integrating AI into a traditional bank is no easy task. A survey by New Vantage Partners found that 77% of financial services leadership executives identified AI adoption as a significant challenge, with people and processes listed as the main factors causing difficulty. Couple this with emerging regulatory scrutiny – how should banks approach AI, balancing risk appetite with the ambition to leverage powerful use cases?

Critical to success is developing a strategy and operating model that is ‘AI-ready’. An AI-ready bank requires new structures, skills, processes, governance, data flows, and fundamentally, a new AI strategy. There are five key ways to achieve this:

1. Define your AI strategy

An AI strategy is an ongoing journey, not a one-time project; and it’s something many banks have already begun to define. A good AI strategy is aligned with the wider enterprise strategy and objectives, while being specific about the steps needed to achieve the desired use of AI.

To develop an AI strategy, or test an existing one, traditional banks should ask themselves several questions: To what extent is our market being disrupted by new technologies, and how do we need to respond? What use cases are we interested in exploring? How do these align to our enterprise strategy? What is our current maturity in terms of AI capabilities? What is our appetite for risk? Do we see ourselves as a first mover, or would we rather wait for others to successfully pilot initiatives? What is our investment appetite?

2. Create an AI studio

At the heart of AI-enabled banks is the ‘AI studio’. This is a central team responsible for identifying AI applications and developing them into fit-for-purpose use cases. Yet this team’s role extends beyond development; it also nurtures the implementation of AI across the bank, dictating how to accommodate changes to structures, skills, and processes. And crucially, it provides governance around the ethics of AI, whether its users are employees or customers.

Traditional banks would do well to create an AI studio that can empower and upskill employees to leverage new AI tools and engage stakeholders with innovations – all while shaping and refining AI use cases and providing security through AI framework policies.

3. Be adaptive

AI is a technology that disrupts like no technology before and, as a priority, banks will need to adapt more than ever before. An adaptive organisation can continuously sense why, when and where it needs to introduce AI, and respond by pivoting resources appropriately.

What this means in practice is that business line leads, HR and the AI studio need to work in synchrony. The AI studio detects capabilities that can be fulfilled by AI; HR ensures that via strategic workforce planning, impacted roles can be redefined while continuing to leverage the strengths of individuals; business leaders support and guide the change, ensuring the workforce retains a sense of empowerment.

4. Optimise AI through data and analytics

Banks’ successful implementation of AI models also hinges on reliable, accurate and unbiased data. A robust analytics function is also needed to facilitate AI model training and optimisation.

To ensure AI systems provide reliable outputs, banks must also establish frameworks that promote data quality, diversity, and privacy. This involves adopting data cleaning and validation processes to uphold data quality.

While this will be largely overseen by IT and data management teams, active involvement from other employees is vital to ensure data is tailored to their needs, and to gain buy-in to the policies governing AI’s use.

5. Regulate AI through leadership and governance

The challenges and opportunities that AI affords can at times feel overwhelming for leaders, as evidenced by a recent edX survey of C-suite executives. But ultimately, leaders should treat AI no differently to any large-scale technology-led transformation, such as adoption of enterprise applications or the move to cloud.

Effective AI leadership involves understanding, at a high level, AI’s capabilities and limitations; identifying where it can create value in line with business strategy; and encouraging a culture of AI adoption to gain a competitive advantage.

At the same time, there is a need for an AI representative at leadership level to offer expert insights and act as a single point of accountability. This could be a designated Chief AI Officer, or an existing role (such as Chief Digital Officer) with an appropriately informed and passionate individual.

Fully integrating AI governance with wider organisational governance is also key. The Risk Committee must interrogate questions of transparency, equity, and security. The Finance Committee approves spend in line with strategic business objectives. And the IT Committee ensures that AI R&D and implementation aligns with overall technology strategy.

No two banks will have the same approach to AI, and hence each vision will be unique. Whatever the future of banking, it is increasingly clear that the winners will get there through the power of AI. Any efforts spent ensuring your bank’s operating model is AI-ready through undertaking these five key measures will quickly pay dividends.

This article was first published in Finance Derivative 

Explore more

Contact the team

We look forward to hearing from you.

Get actionable insight straight to your inbox via our monthly newsletter.