David Rowlands, global head of AI at KPMG for the past 18 months, has been tasked with prioritising AI within the professional services company, as well as guiding clients on AI strategies that often require rethinking entire operating models.
Previously, he led KPMG’s UK consulting practice for eight years. Rowlands speaks exclusively to TI to share his insights on how AI is transforming industries, the challenges organisations face, and what the future holds.
Why did KPMG create a dedicated AI leadership role?
A chief AI officer needs to take a broad view of AI’s impact on an organisation. It’s not just about technology; it’s about reimagining business structures. My background in leading one of our biggest markets helps me understand where AI can take us.
At KPMG, my role involves integrating AI into how we go to market, our sales processes, and enhancing our audit, tax, and advisory solutions. We’re also rethinking our delivery model and preparing for emerging technologies like quantum computing.
What’s one of the biggest challenges in your role?
Ensuring people don’t ‘outsource’ the AI problem. It’s about equipping everyone with the skills to use AI safely and effectively. For instance, we launched ‘24 Hours of AI,’ a global training initiative accessible to all 275,000 KPMG employees. The message was clear: AI is for everyone.
How are your clients feeling about AI?
There’s a mix of optimism and anxiety. Many CEOs see AI as crucial for long-term success, but they’re also aware of the potential disruption. We’re at a paradoxical moment of peak hype and peak hesitation due to implementation challenges.
What categories of AI are most relevant to your clients?
We see four main categories:
Role Augmentation: Enhancing individual productivity without changing existing roles. This has delivered efficiency but within traditional structures.
Agents: This is where real transformation begins. AI agents, powered by curated knowledge bases, are evolving from decision-support tools to autonomous decision-makers. We’ve implemented this in a supermarket call centre, where AI suggests actions based on policy databases, improving call quality, reducing costs, and enhancing customer satisfaction.
AI-Centric Operating Models: Organisations are beginning to rethink their entire structure. We help clients, like an American bank, redesign their operating models using AI, aligning them with strategic business cases for exponential growth.
Cognitive Engines: These address complex, sector-wide problems. In healthcare, for instance, AI helps the NHS optimise resource allocation. In banking, AI is transforming credit risk policy management, a billion-dollar challenge.
Where is AI currently driving the most efficiencies?
Mostly in cost-saving and efficiency gains by enhancing existing processes. But soon, AI will eliminate outdated processes altogether. Organisations must experiment, learn, and collaborate with technology partners to stay ahead.
What mistakes do companies make when implementing AI?
Many get stuck in experimentation without achieving scale or ROI because of data challenges, cloud infrastructure complexities, and misunderstood costs.
Remember that first bill that organisations got for their cloud compute? It can be quite difficult to understand what you’re paying for. It’s important to build in telemetry and measurement of the usage so that companies can start to unpick their costs and figure out what the cost drivers are.
PwC: Processing insurance claims with AI and human oversight
Part of the issue is structure: Because AI often optimises existing structures rather than reimagining them it can be quite hard to make sure that you’re capturing where the benefits are going, especially if you are still optimising the ‘old world’ using AI rather than challenging its very structure.
AI changes tasks, roles, and organisational structures. Businesses must rethink roles, reskill employees, and maintain trust.
Setting the right pace of transformation for their business is another area where companies need to think carefully. While some are existentially driven by technology – others will adopt AI through technology providers such as Oracle, Microsoft, Salesforce, who are introducing AI woven into the fabric of their offerings.
For the vast majority, this is going to be their AI journey.
Over a decade ago companies restructured for digital, with many evolving into content owners. Will the uptake of AI see organisations follow a similar path?
Dusting down some of the thinking on this is not a bad idea, because, again, content is massively important in the world of AI, and who owns that content, and how do you protect it? That is still important. So that’s not a bad paradigm.
AI is creating new business opportunities, but businesses will also need to rethink operating models, human roles, and governance for an AI-centric future.
What about data readiness? How crucial is that?
AI consumes data rapidly, so having ‘AI-ready’ data is vital. We see data as a pyramid: core data, insight, knowledge, and wisdom. Protecting and using knowledge as data is critical.
It would be a mistake to delay your AI journey until you’ve put all your data in a structured data lake however, but making sure that you’re investing in your data capabilities in line with your AI journey is really important.
How does KPMG ensure responsible AI?
We’ve embedded a “Trusted AI” framework across all AI initiatives. It’s also offered to clients to help them master responsible AI. This approach ensures consistent governance, ethical usage, and trust.
It’s also another area where the chief AI officer is the only one with a broad-based perspective on an organisation who can put that in place.
How do you measure AI’s ROI and economic impact?
We use advanced tools to evaluate AI’s value case and a complex workforce model to bridge current and future roles. Tracking usage and costs is crucial to understanding AI’s economic drivers.
What’s next for AI at KPMG?
We’re focusing on AI’s role in complex problem-solving, expanding AI-centric operating models, and deepening trusted AI governance. As AI continues to evolve, our mission is to help clients harness its potential responsibly and strategically.
