In a recent episode of the Cambridge Executive Business Insights: Rethinking AI podcast, Professor Jaideep Prabhu hosted David Ferne, Vice President of AI at NTT Data Services, for an exploration of how companies can move beyond surface-level applications toward purposeful, scalable AI strategies that unlock broader organisational intelligence and creativity.
Today’s business leaders must ask: what does it mean to design organisations, and work, for a smarter, more adaptive AI future? The episode calls for a reimagining of not only the technology but the very ways we organise, learn and lead.
AI’s journey: from hype cycles to human symbiosis
Every wave of AI enthusiasm brings with it both breakthroughs and cautionary tales. For David Ferne, this has been a personal journey. He describes a career traversing both the “summers and winters” of AI, that is, periods of excitement and apparent dormancy, with his first major turning point coming at a 2021 Microsoft conference. There, a hushed invitation to view an early version of OpenAI’s GPT-3 marked, for him, a watershed moment.
What set GPT-3 apart from the natural language technologies that came before was not merely technical scale but the capacity for nuanced, near-human interaction. Having spent years wrestling with brittle, inflexible natural language processing (NLP) systems, David recalls being stunned by the model’s fluency, adaptability and potential to “do more with less”. It was, in effect, AI that could engage with information, not just data, in a manner previously out of reach.
This encounter shaped his approach at Cognizant and, more recently, at NTT Data Services, where the challenge has become not just technical deployment but wide organisational change: what David describes as the effort to “redefine the social contract” between humans and intelligent machines.
Making AI a business strategy, not a bolt-on
Too many organisations, David cautions, still regard AI as a technical product: such as a plug-in, an app, or a procurement exercise, instead of seeing it as a living part of the business. The result has often been fragmented pilots, isolated proofs-of-concept, and a lack of cohesive transformation. Echoing the way city design is most effective when integrated into the wider urban fabric, he argues that true progress only emerges when firms “bake the AI strategy into the business strategy”, which allows both to evolve together.
The pattern of success, then, is clear. The most forward-thinking organisations don’t set up an “AI strategy” siloed from their overarching plans. Instead, they use AI as an enabler, a catalyst for reimagining products, services and even internal culture. Firms that reap the most benefit see AI not as a ticket to efficiency alone, but as a force for growth: innovation that expands possibilities for their workforce rather than merely shaving costs.
This mindset is critical. “Communicate growth first AI initiatives,” advises David, noting that framing AI as a route to efficiency almost inevitably triggers resistance and fear among employees. By positioning AI as a growth tool, firms enable employees to engage with curiosity, to see themselves as partners in a journey rather than as replaceable cogs.
The leadership imperative: healthy scepticism and distributed ownership
Leaders play a pivotal role in determining whether AI becomes a source of transformation or of anxiety. David Fern distinguishes between emotional scepticism (“let’s not touch this, it’s too risky”) and healthy scepticism – the latter being a pragmatic, thoughtful approach that weighs potential with appropriate caution. The leaders who succeed embed good governance, insist on auditability and encourage experimentation within clear bounds.
But modern leadership, he argues, is not just about the C-suite. The coming era will see “AI leadership” diffused throughout the organisation. Every employee, not just a select technical elite, will need to become comfortable managing teams of AI agents, distributing tasks and coordinating workflows that blend machine and human input. In effect, the skill of “managing the collective” will become as important for frontline workers as it is for traditional managers.
Bridging the void: from pilots to production
The gap between aspiration and reality is perhaps nowhere more visible than in the so-called “pilot to production” chasm. Firms can now swiftly spin up AI pilots in days or weeks, but most struggle to translate those exciting prototypes into reliable, enterprise-grade systems. The technical novelty of today’s AI, with its low barrier to entry, can create a false sense of ease: just as building a parklet is easier than reimagining a city’s transportation, deploying one chatbot does not equal systemic change.
To cross this chasm, firms must build the right underlying platforms, robust governance structures and the “trust layer” that enables broad adoption. “[Trust] is the only enduring moat,” David explains, without it, even the cleverest systems will languish unused. Building this trust involves not just technical measures (audit trails, clear accountability, regulatory compliance), but also transparency and engagement with staff.
The myth of data quality: why information matters more
A familiar refrain in digital transformation circles is the lament over “poor data quality.” Firms fear they cannot embrace advanced AI unless their vast, disparate data stores have been fully cleansed, integrated and standardised. Yet David provocatively suggests this conventional wisdom is outdated (or at least incomplete) in the age of generative AI and large language models.
“Data doesn’t matter. AIs don’t use data, they use information,” he asserts, shifting the focus from raw tables of numbers to the so-called “semantic layer”, the structured, meaningful knowledge extracted from the daily flow of decisions, conversations and actions. Most organisations, he argues, already possess this informational substrate: they make complex choices, price products, and navigate ambiguity every day, even with spotty data. It is this context, not mere data, that AI needs to augment.
In the long run, the “gold” for firms may well be found not in their historic datasets, but in capturing the iterative, collaborative interactions between humans and AI agents as they co-create solutions. The insights gleaned in these exchanges such as heuristics, playbooks, creative “what ifs”, could form a kind of enterprise intelligence, a priceless new resource for adaptation and learning.
Governance as an accelerator, not a handbrake
At the heart of the responsible AI debate lies the question of governance. Conventional wisdom imagines compliance and oversight as drags on pace and innovation, as necessary evils to be applied after the fun of prototyping is done. Ferne turns this logic on its head, proposing instead that governance is “an accelerator” when designed into AI initiatives from day one.
Applying governance retroactively, he notes, often leads to a collapse in the value of proof-of-concept AI, as innovations must be refitted into risk frameworks they were never designed to meet. In contrast, building with governance as a “first class citizen” provides the rails for safe experimentation, catalysing innovation by clearly laying out the “rules of the road.” It is a foundation, not a fence.
Yet one thorny issue remains: accountability. As algorithms make more decisions affecting customers, products and the public, firms must grapple with the question: who is responsible? The current “human-in-the-loop” paradigm strains under the complexity and autonomy of modern AI. Without clear lines of accountability, true trust and thus, widespread adoption, will remain elusive.
Curiosity, creativity and the future of work
As AI continues to mature, the most important shift may not be in the technology, but in the culture of organisations themselves. David dreams of a workplace where an employee’s value is measured not by their mastery of obscure software interfaces, but by their ability to ask the most illuminating questions. “Curiosity will be the single most important skill for the future workforce,” he maintains.
In such an environment, technical knowledge and domain experience count for a great deal, not because they enable more efficient button-clicking, but because they equip people to interrogate problems, imagine alternatives and shape AI’s outputs with creative flair. The tools will take care of the low-level execution; it will be the uniquely human capacity for curiosity, judgment and contextual understanding that confers enduring value.
Organisationally, this demands both top-down and bottom-up transformation. Leaders must set the agenda: “the tone is set by the leaders,” as David notes, but they must also foster a culture of distributed experimentation, empowering employees at every level to shape workflows, try new approaches and co-create with AI. Managing the balance, that is, building structures that both govern and encourage curiosity, will distinguish the “destination employers” of the next decade.
Destination employers and the new value proposition
What will make an employer attractive in the AI age? It will not be mere access to the latest tools, nor rigid certification in a particular platform. The future, David suggests, lies in organisations that develop “AI factories”, structures that nurture curiosity and creativity, moving seamlessly from experimentation to production.
As AI takes over more of the technical heavy lifting, the premium on diverse experiences may rise. Firms may need to encourage staff to broaden their perspectives, to travel, to learn from art and science alike, bringing fresh insight to the problems at hand. In this model, creativity and curiosity are not ancillary to productivity; they are its foundation.
Personal agents and the coming transformation
Looking even further ahead, David Ferne notes “the personal agent revolution” as the next major wave. Whereas much AI today is “data centre” based, operating centrally, subject to enterprise control and governance, the near future will see a proliferation of AI “personal agents” acting on behalf of individuals, running on personal devices, managing mundane tasks, synthesising information, and interfacing with larger systems as needed.
Firms will need strategies for both modalities: robust, centralised agents that govern critical workflows and data, and distributed, highly personalised agents that empower workers directly. Productivity gains will come, not from a single killer app, but from seamless interplay between these modes, all anchored in governance and trust.
The call to action: design for intelligence, design for people
Business leaders today must recognise that AI is not simply a technology. It is a catalyst for reimagining how work is done, how teams organise, and how human intelligence is multiplied, not replaced.
The key is intentionality: “be intentional and strategic about AI right from the get go,” urges Professor Prabhu. Leaders must frame technology as part of a long-term journey, grounded in growth, responsibility and trust. They must think deeply about governance and risk, but also about the curiosity and creativity they wish to encourage.
The future of AI-enabled business is not about building bigger machines, but fostering greater capacity for adaptability, learning and transformation in the people and organisations that use them.
To hear the full conversation and explore these themes in greater depth, listen to the full episode, available wherever you get your podcasts.
