Ehisuoria Aigbogun

I support View B — organizations should preserve collaborative problem-solving even when AI can identify solutions faster. While AI can significantly improve speed, efficiency, and root-cause analysis, the value of collaborative problem-solving extends far beyond simply finding the “best” answer. Team discussions build organizational learning, cross-functional alignment, accountability, and long-term capability development. If organizations rely too heavily on AI-generated solutions, employees may gradually lose the ability to critically analyze problems, challenge assumptions, and innovate independently. At John Deere, collaborative problem-solving is especially important during large operational transformations involving multiple systems such as Combine harvester. AI may quickly identify technical root causes or recommend process improvements, but operational reality or design change is often more complex than the data alone can capture. For example, during large system integration and process standardization initiatives, AI tools may recommend the fastest or most statistically effective solution based on historical performance data. However, when the X9 Combine was built, there was a process improvement opportunity with dust impacting visuals of operator's yield when harvesting. I had to lead Cross-functional workshops frequently to uncover hidden dependencies, customer-impact risks, and design gaps that automated systems would have overlooked. We came up with a dust fan system that pulls the dust from the front of the Combine and expels it by the side of the Combine, and we fine-tuned the speed of the fan after multiple field tests under various environmental conditions to account for grain loss. During the course of this project, all stakeholders were included. Engineering (electrical, hydraulic, mechanical), sales, product, software development, and the software systems team. This was rolled out and was a huge success till date. I have seen situations where projects that appeared technically successful created downstream operational challenges because certain user groups were not involved early enough in the decision-making process. In many cases, the issue was not the quality of the AI recommendation itself, but the absence of collaborative engagement that could have identified implementation risks earlier. In my opinion, AI should enhance collaborative problem-solving rather than replace it. Organizations should use AI to accelerate data analysis, identify patterns, and narrow potential solutions, while still allowing teams to validate assumptions, share operational knowledge, and build collective ownership of outcomes. A company that prioritizes only speed may solve today’s problems faster, but over time it risks weakening employee engagement, innovation capability, and organizational learning. Sustainable operational excellence comes not only from having the right answer, but also from building teams that understand why the answer works and how to improve it in the future.

I support View B — Continue the project despite the AI warning. AI predictions should inform decisions, not determine them. Large transformation initiatives often look inefficient and unstable before benefits appear because truly disruptive ideas rarely resemble historical success patterns. If organizations terminate projects too early based on predictive models, they risk optimizing for short-term certainty instead of long-term advantage. A strong example is the iPhone. Before launch, the idea of a phone that was “just a screen” was widely criticized. In 2007, then-CEO of Microsoft, Steve Ballmer, publicly mocked the iPhone, arguing that a touchscreen phone without a physical keyboard would not appeal to business users and question its pricing and market fit. From a predictive model trained on existing mobile phone success factors at the time, battery life, physical keyboards, enterprise adoption, manufacturing certainty. The iPhone likely would have triggered warning signals. Yet Steve Jobs continued despite skepticism. The result redefined not just phones but entire consumer ecosystems. Another powerful example is Nvidia In the late 1990s and early 2000s, NVIDIA experienced severe financial pressure. One critical turning point came when Sony selected NVIDIA hardware for the PlayStation 3, providing revenue and stability during a period when the company was still proving its long-term direction. At the time, NVIDIA’s continued investment in GPU computing could have looked inefficient from a project analytics perspective, high R&D spending, uncertain adjacent markets, and no immediate business case beyond graphics. Years later, the breakthrough came from an unexpected direction: researchers and PhD students began using GPUs for machine learning because GPUs could perform massively parallel computation faster than CPUs. A landmark moment came when researchers demonstrated dramatic reductions in neural network training times and error rates using GPUs for deep learning workloads. What looked like a graphics company’s expensive side bet became the infrastructure layer for modern AI. If an AI project-monitoring system had evaluated NVIDIA in the early years using only historical ROI, cost efficiency, and milestone predictability, it may have recommended cutting GPU compute initiatives and doubling down only on gaming graphics. That would have killed one of the most valuable strategic bets in modern technology. AI should therefore trigger intervention, redesign, and governance review not automatic cancellation. Some of the highest-value projects produce signals that resemble failure before they become breakthroughs.

I support View B. Organizations should distribute opportunities more broadly instead of allowing AI to repeatedly assign critical work to the same top performers. While AI can accurately identify employees with the highest probability of immediate success, leadership is not only about optimizing short-term productivity. Organizations must also build long-term capability, resilience, and future leadership pipelines. If important projects are continuously concentrated among a small group, the company risks creating operational dependency on a few individuals while limiting the growth of the broader team. I work for Dell computers and I can say this is especially important in large operational and transformation environments. During major digital transformation initiatives, organizations often rely heavily on experienced high performers because they deliver faster and reduce operational risk. However, if only the same individuals handle customer escalations, strategic presentations, and transformation projects, other employees never gain the exposure needed to develop problem-solving and leadership skills. For example, Dell recently completed a major database consolidation project in collaboration with Deloitte called “One Dell Way,” which aimed to consolidate tools across the company into a single database. Management selected high-performing employees to evaluate multiple tools and build the new system. One drawback, however, was that many employees who used some of the downstream tools were not consulted or included in the process, and several critical attributes were missed at the May 1 launch. I believe this highlights a significant failure because, beyond limiting employee growth opportunities, it also created a knowledge-utilization problem that increased operational workload. I am currently working with affected employees to gather their requirements and migrate the missing attributes under constrained timelines. In my opinion, AI should support managers by identifying top performers, but managers should still intentionally rotate opportunities across capable employees. This helps distribute knowledge more broadly across teams while also giving employees equal opportunities to develop and grow within the organization. A balanced structure could involve assigning experienced employees as project leads while allowing developing employees to own portions of the project under supervision. This approach maintains operational quality while strengthening the organization’s long-term capability. A company that only optimizes for today’s performance may operate efficiently in the short term, but over time it risks weakening innovation, employee motivation, succession planning, and organizational resilience.

A good example is Human Resources. In my organization, it was gathered, we have over 20 years of documentation that is not organized, but if harmonized and standardized, we can build an AI solution that reduces resource burden from the data management team. A standardized database can help create an Agentic AI solution that can help address inquiry management quickly and only escalate if the resource does not exist in the database.

AI solution is becoming outdated or fragile when there is a decline in the accuracy of response or a second ticket is created for the same issue due to lack of credible answer. Another indicator will be the increasing number of escalations by the solution. Ensuring a long term sustainability will be to continually review the AI logic and train the solution with more data. We could also get subject matter experts to review and audit ensuring the alignment with the business case and user needs.

Slices of Cheese (Defense Layers) in My Standardizing and Automating Server Configuration Database Standardized Development Protocols Clearly defined attributes and terminology leads to standard and consistent communication path between tools and reduces attribute conflict during automation frameworks. This will serve as the first line of defense. They ensure consistency and minimize ad hoc implementations that could introduce variability or error. An example in my current project is ensuring platform names are consistent across multiple teams. Validating the data My current project is ensuring no missing attribute or required data before moving to publish data in production site for engineering utilization. Governance and Documentation Standards I have gone ahead to implement governing policy to ensure the right person gets the right level of access needed. This will align with Dell’s internal policies governing data usage and tool integration offering structure and compliance, particularly around ethical AI concerns. Holes (Weaknesses) in the Process Inconsistent Adoption of Standards Not all teams may apply standardized practices equally, especially across business units, leading to process fragmentation. Poor Change Management Process Introducing automation without proper training of stakeholders or alignment with stakeholders may result in improper use or resistance. Guiding Improvement with Business Excellence Principles Continuous Improvement: Regularly reviewing and refining automation processes, ensuring lessons learned are systematically captured and reused. Customer Focus: Ensure to standardize and design the tools by engaging all stakeholders to ensure we have the right set of information and alignment at every stage of the project. Process Integration: Use Lean Six Sigma tools to identify bottleneck in the process and variability in our automation lifecycle. Lastly, eliminating human/manual data entry into configuration tool. Cultural Alignment: Fostering a culture of quality and accountability where teams embrace standardization and automated process.

I would approach this problem using the DMAIC approach. Understanding the different stores, products and timeline. 1. With these, I would need to validate the issue and understand the impact using data, Example is sales down by 10% in stores with 16% stockouts. Categorize the data by stores, products and time. 2. After completing process 1. This would lead me to ask the right questions which is my root cause analysis. Using the 5 whys approach, Why are sales dropping? Stockouts. Why stockouts? Inventory not replenished. Why not replenished? Forecast underestimated demand. Why was demand underestimated? Forecast model doesn’t account for promotions. Why? Forecasting model lacks integration with marketing calendar. I would use Fish bone diagram to categorize potiential causes across. This will help me explore all possible causes before diving into a specific cause. Example of potential cause in this case could be: Forecasting methods, machines, environment, seasonality. 3. Next step is to run a pareto analysis to find the impacted categories. Additonally, I will run a correlation study to validate the relationship between forecast accuracy and stockouts. I can track trend analysis which can show if stockouts is due to seasonality of if this is event based. 4. While going through these process, I will ensure my data collection process is unbiased or there are no errors in data collection process. Additionally, I can collects anecdotes to track correlation between the anecdotes and the data. 5. Lastly, I will prioritize root cause based on correlation with the problem and how they can be controller.