No single road: why harnessing AI is a nuanced business

Now that digital transformation has become a business imperative, the rise of generative AI (GenAI) is reshaping industry norms and creating new models for success. However, with GenAI sitting at the pinnacle of Gartner’s Hype Cycle, the technology is at risk of being turned into a corporate panacea. Some companies may see it as a solution to almost everything, while others may be quick to rebrand their more traditional tech offerings as being “AI-driven”. Yet looking beyond the hype, companies in various sectors are successfully navigating the generative AI adoption journey by balancing innovation with their operational realities.

The digital divide

An analysis of businesses adopting generative AI throws up some stark contrasts. On the one hand, there are digitally native companies, such as ecommerce and travel giants, moving rapidly to incorporate AI-driven solutions into their operations. These sectors have the kind of agile infrastructure that enables them to seamlessly integrate advanced technologies, such as chatbots and recommendation engines, enhancing customer experience and operational efficiency.

On the other hand, traditional sectors, including banking and insurance, are taking a more measured approach. The integration of generative AI in these sectors is more complex, owing to legacy systems and stringent regulatory requirements. This disparity in adoption rates raises questions about the strategic deployment of AI technologies across varied industries.

Banking and insurance: proceeding with caution

According to S&P Global, in sectors such as banking and insurance, where companies fight hard to differentiate themselves from rival businesses, and where margins are slim, generative AI has great potential to offer a competitive edge. These industries are eyeing GenAI for applications ranging from fraud detection and legacy migration to customer service enhancement. However, the path to implementation throws up plenty of challenges.

Kshitij Jain, EMEA Practice Head & Global Chief Strategy Officer, Analytics at EXL, points out that the optimisation problem for generative AI in financial services is multifaceted. It involves balancing parameters such as business value, accuracy, compliance, user experience and cost. "In the banking sector, the potential for value creation is immense, due to the large customer base and digitised processes. However, the concern for compliance in these heavily regulated institutions cannot be overstated," he says.

He points out that smaller, less regulated financial institutions can be more agile in adopting AI, especially in areas such as product development. Understandably, larger banks exhibit greater caution, often focusing AI applications on enhancing employee productivity, internal processes and legacy code migration rather than direct customer engagement.

Technology with humanity

Digital-first sectors such as ecommerce and travel are in the vanguard in embracing generative AI. Their modern infrastructures enable them to integrate AI more smoothly and to swiftly harness its full potential. For these industries, generative AI isn’t a futuristic concept but a present-day tool driving customer engagement and business growth.

While the technical aspects of AI integration are rightly celebrated for the efficiencies they bring, the human element remains pivotal. Will Hyams, Group Head of Data at global insurance group Howden, emphasises the importance of aligning AI initiatives with people-centric values. "Our goal is not to replace our quality service with impersonal automation but to enhance it, making our teams more informed and efficient," he explains. This perspective underlines the necessity of balancing technological advancement with maintaining human interaction, especially in client-facing industries.

The transition to generative AI comes with a particular set of challenges. There are ethical questions and risks associated with AI technologies, and Hyams stresses the need for human input to ensure model reliability. Data security, quality and compliance with privacy regulations are other areas that demand attention.

For companies hesitant about embracing generative AI, Jain advises a holistic approach that considers the entire ecosystem of legal, compliance, data and technology architecture. "Generative AI is not a one-person job. It requires collective effort and strategic thinking across various organisational functions," he says. Indeed, part of his role at EXL is about helping to ensure that its customers are “AI-ready”, in the sense that they must have well-governed data to feed into LLM programmes, supported by guardrails for responsible AI, model management and prompt engineering.

“Otherwise, it’s ‘garbage in, garbage out’,” says Jain. “And you need to keep enhancing model and data governance frameworks to meet ever evolving regulatory requirements.. As new forms of AI emerge, consuming much more data across many more use cases, that need for monitoring and governance will keep accelerating.”

AI in the boardroom

What’s exciting, say experts, is the shift from AI use cases being predominantly centred on back-office tasks to its potential as a strategic partner in boardroom decisions – particularly in scenario modelling. In a recent discussion, Yuval Atsmon – a senior partner who leads the new McKinsey Center for Strategy Innovation – said that using AI in business strategy could significantly improve outcomes. “Executives shouldn’t fully rely on predictive AI, but it provides another systematic viewpoint in the room,” he says.

Successful adoption comes, it seems, when AI initiatives are rooted in concrete use cases and integrated into the fabric of the business. As the business world stands at the cusp of a generative AI revolution, the journey is as diverse as the sectors embarking upon it: from cautious steps in banking and insurance to rapid strides in digital-first industries, the adoption of AI must be a nuanced, strategic decision. Understanding the unique challenges and potentials of each sector, coupled with a holistic approach encompassing all facets of the organisation, will be key to harnessing the transformative power of generative AI.