Dr. Babajide Ojuola, Executive Director Technical Services, International Energy Services Limited.gettyIn my previous article, I argued that successful AI transformation is fundamentally a leadership challenge rather than a technology challenge. The next competitive advantage in enterprise AI will not come from larger language models or more sophisticated algorithms alone. It will come from organizations that learn how to combine AI with human knowledge.Generative AI has transformed what organizations believe technology can accomplish. Yet despite unprecedented investment, enterprise-wide adoption in engineering companies continues to face significant barriers. The most significant barrier here is organizational trust. Beneath that trust challenge lies an even deeper issue: the absence of effective knowledge management systems capable of preserving, validating and applying institutional knowledge alongside AI.Engineering organizations need more than powerful AI models. They need an operating framework that integrates AI with human expertise. That is why I propose an Assistive Human-Centric AI Knowledge Management Framework (AHCAI-KMF) for enterprise AI in engineering organizations.I define AHCAI-KMF as an enterprise operating model in which AI augments rather than replaces human judgment. It combines tacit knowledge, explicit knowledge, engineering expertise, organizational memory and AI-enabled analysis to improve decision quality, preserve institutional knowledge, enhance organizational agility and accelerate learning. Rather than positioning AI as a substitute for experienced professionals, the framework treats it as an enabler of knowledge creation, knowledge transfer and collective intelligence.I see enterprise AI developing through three complementary capabilities. The first is generative AI, which helps organizations create, summarize and synthesize knowledge from existing information. The second is agentic AI, which enables intelligent systems to plan and execute tasks with increasing levels of autonomy. The third is assistive human-centric AI, which focuses on preserving, validating and continuously enriching the human expertise that underpins complex organizational decisions.These capabilities should not be treated as competing technologies. Generative AI expands what organizations can create and understand. Agentic AI expands what they can execute. Assistive human-centric AI expands what they can learn, retain and improve. Together, they can create an enterprise in which knowledge is continuously generated, intelligently applied and progressively refined through human experience.My nearly 20 years of engineering and project experience has taught me that delivering complex oil and gas and infrastructure projects requires much more than technical compliance. It involves balancing safety, operability, maintainability, constructability, regulation, commercial realities, schedule pressures and technical risk in the ever-changing and uncertain business environment.Many of these decisions depend less on documented standards and procedures than on tacit knowledge accumulated through years of project execution. Two design alternatives may comply with the same engineering codes, yet an experienced project team may recognize that one presents lower lifecycle risk, fewer constructability challenges, greater operational flexibility or improved maintainability.The same form of judgment is required during HAZOP studies, constructability reviews, commissioning readiness assessments, brownfield modifications, shut-down planning, interface management and contractor coordination. Such decisions are rarely resolved through technical documentation alone. They depend on experience, mentorship, professional intuition and lessons accumulated across multiple projects.This tacit knowledge is one of an engineering organization’s most valuable assets, yet it is often among the least protected. It can disappear when experienced employees retire, change roles, leave the organization or complete a project without systematically transferring what they have learned.Engineering also differs from many other knowledge-intensive sectors because its decisions may carry safety, operational, environmental and financial consequences for decades. A choice made during front-end engineering design can influence an asset’s constructability, operating cost, reliability and safety throughout its lifecycle. Preserving engineering judgment is therefore not merely a productivity initiative. It is a strategic risk management capability.Philosopher Michael Polanyi famously observed that “we know more than we can tell.” Few statements better capture the enterprise AI challenge. Generative AI can process documented knowledge, but engineering excellence frequently depends on knowledge that has never been fully documented. AHCAI-KMF is designed to help organizations capture and amplify this expertise rather than allowing it to remain inaccessible or disappear.Nonaka and Takeuchi’s SECI model offers a useful foundation for understanding this process. Organizations create knowledge through socialization, externalization, combination and internalization. Generative AI can significantly accelerate combination by processing and synthesizing explicit knowledge. Assistive human-centric AI can support the full knowledge cycle by helping experts externalize experience, combine it with AI-generated insights and internalize improved practices across teams.Viewed through this lens, AHCAI-KMF is not another AI model. It is the enterprise operating layer connecting people, knowledge and intelligent systems. It helps ensure that engineering decisions benefit not only from historical records and AI-generated recommendations, but also from the accumulated wisdom of experienced professionals. Each project then becomes more than a delivery exercise; it becomes an opportunity to strengthen the organization’s collective intelligence for every project that follows.Leading research increasingly reinforces this perspective. Studies on enterprise AI adoption consistently identify governance, workforce capability, organizational readiness, explainability and trust as important determinants of successful scaling. Technology enables AI, knowledge enables collective intelligence and trust enables enterprise scale.The implications extend beyond engineering. In every knowledge-intensive sector, competitive advantage will increasingly depend on creating environments where AI strengthens human expertise, preserves institutional memory and improves decision-making. Combined with digital twins, enterprise knowledge repositories and structured lessons-learned systems, assistive human-centric AI can help transform engineering companies into continuously learning organizations.AI is ultimately less an automation challenge than a knowledge management challenge, particularly in engineering project delivery. Generative AI is transforming how organizations create knowledge. Agentic AI is redefining how work can be executed. Assistive human-centric AI ensures that organizations preserve, validate and continuously improve the wisdom that guides both.Engineering has always been a discipline of accumulated knowledge. Every successful project leaves behind lessons that should make the next project safer, faster and better. Enterprise AI should accelerate that cycle, not interrupt it.The organizations that lead the next decade will not necessarily be those with the smartest algorithms. They will be those that build the smartest learning systems—where AI amplifies human intelligence, institutional knowledge becomes strategic infrastructure and every project strengthens organizational wisdom.​Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?