Discussion Board 2.3: Think & Share💡 AI Transformation or AI Hype?

I partially agree with the statement. I believe we should distinguish between Generative AI as a support tool and Agentic AI that is delegated responsibility. The adoption of Generative AI to assist every employee should be immediate. Giving employees access to AI is similar to giving them an encyclopedia, a highly capable research assistant, or a knowledgeable colleague available at any time. The productivity gains are immediate, while the risks remain relatively low because the human is still making the final decision. The conversation changes when organisations start assigning responsibilities to agentic AI systems. Once AI is allowed to make decisions, trigger actions, or interact autonomously with customers or other systems, governance, accountability, and ethical considerations become critical. In banking, for example, I see enormous value in AI supporting credit analysts by gathering information, monitoring news, identifying red or green flags, summarising financial statements, and highlighting inconsistencies. This could dramatically improve both productivity and the quality of risk analysis. However, I believe the final credit decision—and the responsibility for that decision—should remain with a human professional. Someone must validate the AI's conclusions and ultimately be accountable for any mistakes. The organisations highlighted in this module, such as Morgan Stanley and Oscar, are successful because they use AI primarily to augment human judgement rather than replace it. That, in my view, is the right approach today. Leaders should therefore encourage rapid experimentation with Generative AI across the workforce, while introducing Agentic AI more gradually, with robust governance and clear accountability. Moving fast is important, but moving responsibly is what will ultimately build trust and create sustainable competitive advantage.

I largely agree that most organisations are adopting AI faster than they are preparing for its operational, ethical, and human implications. The excitement around AI has outpaced organisational readiness. While AI demonstrations and prototypes often create a sense of urgency, many organisations underestimate what it takes to successfully operationalise AI at enterprise scale. A useful analogy is the evolution of household electricity systems. Many homes built in the early 1900s were designed for a handful of appliances. Decades later, when air conditioners became commonplace, the existing electrical infrastructure could not support the increased load. Simply installing more air conditioners without upgrading the wiring would overload the system. The solution was not to stop using air conditioning, but to modernise the underlying infrastructure. I believe enterprise technology is in a similar position today. Most legacy systems, operating models, governance processes, and workforce capabilities were never designed with AI in mind. Organisations cannot simply "plug AI in" and expect transformational outcomes. They first need to rethink their data architecture, governance, business processes, security controls, and employee capabilities. Throughout my experience leading technology transformation, I have seen organisations make two very different choices. In one case, senior leaders became extremely excited after seeing AI prototypes. The immediate conversation shifted from innovation to cost reduction, with assumptions that AI would quickly replace large parts of the workforce. However, they soon realised that a successful prototype is very different from a production-ready enterprise solution. Production systems require robust governance, security, monitoring, compliance, change management, and continuous improvement. The initial optimism was replaced by a more realistic understanding of the complexity involved. In contrast, I have also seen organisations adopt AI responsibly and successfully. Rather than rushing into enterprise-wide deployment, they invested first in AI literacy. Selected employee groups participated in pilot programmes, provided structured feedback, and helped shape broader adoption. AI champions and subject matter experts held regular office hours to support users, while dedicated AI governance teams reviewed use cases, managed risks, and ensured that AI solutions met security and compliance requirements before reaching production. This incremental approach created confidence rather than resistance. The organisation also started with focused, high-value use cases instead of attempting to automate everything at once. Internal knowledge assistants such as AskHR and AskCompliance helped employees retrieve trusted information more efficiently. Similarly, one of my interns developed an AskDynamics assistant to support Microsoft Dynamics users with customer support queries. These narrowly scoped AI assistants delivered measurable value while allowing the organisation to learn, improve governance practices, and build trust. Another important enabler was simplifying the technology landscape before layering AI capabilities on top. By consolidating fragmented customer systems into a single CRM platform, users gained a unified view of customer interactions. AI capabilities such as Copilot, email drafting, and sentiment analysis were then introduced gradually in phases rather than all at once. This phased rollout reduced disruption, improved adoption, and allowed users to build confidence over time. These experiences reinforced an important lesson: responsible AI leadership is not about implementing AI as quickly as possible. It is about creating the organisational conditions that allow AI to succeed sustainably. Technology alone does not deliver transformation. Organisations must invest equally in governance, data quality, employee capability, ethical oversight, and change management. Leaders should ask not only "Can we build this?" but also "Should we build it?", "Are we ready to operate it?", and "How will it impact our people?" Ultimately, organisations that treat AI as a long-term capability rather than a short-term technology project are more likely to realise sustainable business value. Responsible AI adoption requires patience, strong governance, continuous learning, and thoughtful leadership-not simply faster deployment.