The Promise of AI in the Enterprise
By the mid-2010s, artificial intelligence had evolved beyond a research topic into a business imperative. IBM was already a pioneer with its Watson platform and quickly positioned AI not merely as a technology but as a strategic enabler capable of transforming enterprise processes, decision-making, and client interactions. Watson's cognitive computing capabilities, including natural language processing, machine learning, and advanced analytics, were designed to help organizations turn unstructured data into actionable insights while streamlining operations and opening new avenues for innovation.
Watson's promise was enormous, from automating routine workflows to enhancing complex decision-making in industries such as healthcare, finance, and retail, and signaling ten years ago that AI could augment human expertise. The platform offered the potential to accelerate research, detect patterns invisible to traditional analytics, and support frontline employees in making faster, more informed decisions. At the same time, IBM faced the challenge of scaling these capabilities across global clients with varying levels of technological maturity, especially when many organizations were still hesitant to adopt cloud infrastructure. Adoption was uneven, and early deployments often required not only technical integration but also cultural shifts within client organizations.
IBM's approach combined technological rigor with human-centered principles. Cognitive computing was positioned as a partner to human decision-making, emphasizing that collaboration between machines and people was essential for enterprises of all sizes. Training programs, workshops, and digital badges became essential tools, enabling teams to learn iteratively and apply AI thoughtfully in diverse contexts. This dual focus, technological and human, became the hallmark of IBM's AI strategy and a model for enterprise adoption worldwide.
From Service Delivery Manager to AI Advocate
In my role as a Service Delivery Manager (SDM) supporting Delivery Partner Executives (DPEs) in the Netherlands, I had an opportunity to be part of that journey. My teams worked with smaller accounts, designing pilots, crafting proof-of-concept models, and integrating IBM's capabilities into client workflows. These engagements were often structured through multi-year contracts and required flexibility, patience, and the recognition that technology alone could not guarantee success.
Mentorship was critical during this phase. My mentor was a general manager and former artillery instructor in the Netherlands who emphasized operational rigor, client empathy, and strategic foresight. Under his guidance, I was offered responsibility for a team transferred from Belgium as resources moved to Poland. I chose to decline the opportunity because I knew some of the people affected personally and cared deeply about the consequences. This episode underscored a difficult insight: large organizations often make decisions that prioritize transformation and efficiency over long-standing relationships.
While that reality was unavoidable, digital literacy, reskilling, and human-centered design helped create a more balanced approach to AI adoption. IBM encouraged employees to embrace continuous learning, upskill, and participate in workshops that emphasized iterative experimentation. Over time, I transitioned from a learner to a coach, helping colleagues in Poland organize workshops and strengthen their capabilities. In parallel, we supported clients on similar journeys, integrating AI into their processes without disrupting existing operations. This dual commitment to internal development and client success became a cornerstone of our approach.
Scaling Cognitive Solutions Across Industries
Scaling AI solutions across industries required a combination of technical sophistication, critical thinking, and strong governance frameworks. Many of my client accounts initially resisted change, but some gradually adopted an iterative approach. Instead of massive rollouts, teams worked closely with clients to co-create solutions, test them in real-world scenarios, and refine contracts and service models based on feedback. Watson's applications extended across industries, from healthcare systems analyzing medical literature to assist oncologists, to financial institutions detecting anomalies in risk models, and retailers optimizing supply chains. These deployments were complex and required careful adaptation to each organizational context.
Scaling AI years later surfaced structural and ethical challenges. Data quality, algorithmic transparency, and explainability became pressing concerns for clients adopting cognitive systems. Organizations needed to trust AI recommendations, which required rigorous validation, governance frameworks, and human oversight. IBM invested heavily in these practices, embedding them into the operational DNA of new client engagements. Cognitive computing was not a plug-and-play solution but one that required careful integration with processes, policies, and human decision-making structures.
Everywhere, AI adoption illuminated the importance of organizational culture. Teams had to embrace iterative learning, tolerate early failures, and collaborate across disciplines to deliver results. The alignment of leadership, technical expertise, and frontline employees became a non-negotiable factor in success, while human-centered design provided a unifying framework. This approach ensured that new technology implementations could be both technically robust and socially responsible.
Lessons Learned and the Path Forward
The journey of IBM highlights a central principle: technology alone cannot drive transformation. Sustainable change requires infrastructure, well-defined processes, data-driven decision-making, governance, and human engagement. Even sophisticated cognitive systems like Watson only became effective when employees understood the challenges involved, organizational structures supported experimentation, and leadership encouraged continuous learning.
During this period, I transitioned from operational service delivery tasks toward shaping strategy, mentoring teams, and championing enterprise-wide adoption of new capabilities. Human-centered design, mindfulness in decision-making, and iterative experimentation became principles that guided many of my professional choices. These approaches proved critical in embedding AI sustainably into enterprise practices while reinforcing leadership built on empathy, adaptability, and continuous learning.
Digital transformation in enterprises is rarely linear, and cognitive computing proved to be a journey with many setbacks as well as breakthroughs. AI adoption required resilience, patience, and a commitment to thoughtful design across both services and products. The company's work with Watson illustrated how IBM gradually moved away from the early hype and toward a more pragmatic strategy where human expertise was augmented to improve decision-making and innovation across industries.
The Jeopardy match between IBM's Watson and champions Ken Jennings and Brad Rutter in 2011 was heavily engineered. Watson was connected directly to the game's electrical system, allowing it to buzz in within milliseconds of the signal being activated. The system did not "listen" to the presenter's clues but received them in electronic text format for faster processing. The event demonstrated impressive technological capability, but it was not truly equivalent to human competition.
The company later realized that its initial strategy had been too ambitious. Instead of pursuing large "moonshot" initiatives, IBM shifted toward a more grounded platform strategy for enterprise AI. Watsonx emerged as a new cloud-based platform with dedicated infrastructure, governance models, and consulting services designed to support enterprise adoption. At the same time, IBM spun off its lower-margin managed infrastructure services into a new company, Kyndryl, allowing IBM to focus on software, platforms, and AI-powered solutions.
Looking back, lessons resonate and present the constant need to re-evaluate challenges and solutions, with multiple interactions shifting from a minimum viable product to a final service or product, after multiple interactions. Organizations must integrate technology with mindful human judgment, foster organizational agility, and create learning environments that encourage experimentation and reflection. Watson's story demonstrates that AI's transformative potential is unlocked not by code alone, but by thoughtful integration into human-centered enterprise practices, which can turn into a blueprint for sustainable digital transformation.