In every serious transformation, there is a protagonist under pressure, a problem beneath the surface, and a choice about which future to inhabit. In the era of agentic AI, the protagonist is the organisation itself, an enterprise that wants to remain relevant and resilient in a world where knowledge is abundant but wisdom is scarce. The presenting problem is not a lack of AI tools; it is the confusion between tools and true transformation, between an entwined relationship with AI that amplifies human judgment and an entangled one that multiplies noise, anxiety, and stagnation.
On one path, the firm is driven by fear. Leaders see headlines about disruption, watch competitors launch AI initiatives, and feel compelled to “do something with AI.” Pilots proliferate, platforms are procured, demos are showcased. AI appears everywhere, in presentations, dashboards, and internal campaigns, but rarely in the deep work that truly creates value. Automation is bolted onto existing processes without questioning the assumptions beneath them. Teams are handed new tools but not a new narrative. The result is entanglement: fragmented bots, opaque systems, and a growing sense that machines are influencing decisions humans do not fully understand retrofitted onto systems never designed for them. People feel more insecure than empowered, and the firm’s story becomes one of motion without meaningful progress.
On another path, the firm chooses a different stance. Instead of asking, “How do we avoid falling behind?”, leaders ask, “What problems are we uniquely positioned to solve that were previously out of reach?” AI ceases to be a symbol or checklist item and becomes a design material for reimagining how the enterprise senses, thinks, and acts. Human capabilities, judgment, ethics, narrative, courage, are placed at the centre, while AI agents are tasked with scanning environments, reasoning across complexity, and orchestrating actions at a scale no team could manage alone. In this story, human and machine capabilities are deliberately entwined, each clarifying and amplifying the other.
The tension between these two paths is sharpened by a simple reality: knowledge has become cheap. Powerful models can retrieve, summarise, and recombine information from vast sources on demand. What was once rare, access to expertise, market data, technical insight, is now available to any organisation with connectivity and modest resources. Yet decisions do not automatically improve. The bottleneck shifts from acquiring information to discerning what matters: which questions to ask, which signals to trust, which trade‑offs to make, and how to align choices with long‑term values and strategy. Firms either become factories of shallow activity or producers of genuine wisdom.
Fear‑driven adoption almost always leads to the former. Many enterprises follow a familiar pattern: according to a recent MIT report roughly 95% of AI projects stall at pilot stage, creating proofs‑of‑concept and limited wins but never deeply reshaping core processes, structures, or business models. The dominant motivation is defensive (“we must not be seen as lagging”) so the focus falls on visibility rather than integration, learning loops, or governance. Leaders end up confusing the presence of tools with the presence of transformation. People at the front lines sense the disconnect: they see new dashboards and copilots, but little change in how decisions are made or how value is created. Cynicism grows, and “AI” joins the long list of buzzwords that passed through without altering the firm's trajectory.
There is, however, an alternative pattern that some firms have begun to inhabit. In it, leaders recognise that the real work is not to deploy more AI, but to re‑architect how human and AI capabilities work together. They start by clarifying their calling: not only what they sell, but what kind of impact they exist to create in the world. Against that backdrop, they ask how agentic AI as autonomous, goal‑driven, learning systems that can perceive, reason, plan, and act, might help them address previously intractable problems. The conversation shifts from efficiency to purpose, from “how do we automate this task?” to “what new possibilities does this make thinkable?”
Turning that intention into practice requires simple, shared ways of thinking about intelligence in the firm. One powerful pattern revolves around three perspectives on knowledge: how it is searched, how it is managed, and how it is found through people and systems. From the first perspective, AI agents are used to continuously scan markets, scientific developments, operations, and customer behaviour, surfacing weak signals and non‑obvious patterns. Humans decide where to look, which questions to pursue, and what counts as meaningful. Knowledge search becomes a designed capability, not just the sum of individual curiosity. From the second, knowledge is no longer treated as a static warehouse of documents and dashboards. Instead, it becomes a living system that AI organises, summarises, and connects in context, showing relationships and hypotheses rather than just records. Humans interpret, challenge, and embed these insights into strategy, culture, and design choices. Knowledge management evolves from filing to sense‑making. From the third, the organisation develops an extended transactive memory: an awareness of “who knows what, and where to find it” that includes individuals, teams, specialised AI agents, and external partners. AI helps map expertise and route problems to the right combinations of people and systems. Humans define trust boundaries, escalation paths, and ethical constraints so that proactive knowledge deployment extends and amplifies, rather than overwhelms, human agency as the critical differentiator. Together, these perspectives allow leaders to tell a more coherent story about how intelligence flows. Instead of being scattered across tools and departments, sensing and learning become shared, discussable capabilities.
Time adds another critical dimension. When everything feels urgent, it is easy to default to short‑term moves that soothe anxiety but do little to change the underlying trajectory. Yet real change unfolds across different horizons: the systems that dominate today, the transitions already emerging, and the futures that might eventually redefine the field. Firms that remain entangled with AI tend to blur these horizons and become trapped by the inertial lure of doing the familiar faster. They mix up flashy “future” narratives with reactive fixes to immediate problems, letting fashion rather than strategy determine where attention goes. More deliberate firms separate and then reconnect these horizons. They ask, “What are the most important changes we can make in the next 1–2 years to improve how we sense and decide?” alongside, “What possibilities 5–12 years out would we regret not exploring now that we have agentic AI?” They then design bridges as intermediate capabilities, platforms, and ventures that connect present strengths to future opportunities. This provides a shared storyline: where we are, where we could go, and how near‑term choices either open or close long‑term paths.
Concrete examples make the contrast vivid. In some firms, AI is scattered across small experiments: a chatbot in one unit, a forecasting model in another, an automated report somewhere else. Each is measured in isolation, and none has enough weight to alter how the firm thinks or behaves. In others, AI is woven into a holistic experience that touches both operations and customers. Take NIO, a Chinese EV manufacturer that has redefined itself as a lifestyle‑centred “user enterprise,” by embedding agentic AI across in‑product assistants, autonomous services, and a connected ecosystem of digital and physical experiences. In this example, technology is inseparable from the story people live: proactive, personalised interactions that build loyalty through convenience, safety, emotional resonance, and a sense of belonging.
Inside firms, a similar divergence appears in how innovation is approached. One approach treats AI mainly as a way to squeeze more from existing processes, focusing on cost and efficiency. Another recognises that AI can support multiple forms of entrepreneurship at once: refreshing internal capabilities, renewing how the firm competes, creating new offerings and markets, and redefining entire domains. When AI is entangled, these directions are pursued, if at all, through disconnected efforts that compete for resources and attention. When AI is entwined, they become phases of a virtuous cycle. Internal improvements feed new ventures; new ventures generate knowledge that strengthens the core; and over time the firm becomes both more resilient and more exploratory.
This framing raises the stakes of the choice leaders face. Remaining in the fearful, tool‑centric pattern means building ever higher piles of data and models without becoming any wiser. Employees learn to treat each new initiative with scepticism or worse, blind acceptance. Customers sense polish but not depth. The firm spends heavily but drifts, its identity defined by “keeping up” rather than by shaping its space. Committing to the more purposeful path makes a different outcome possible. AI initiatives become chapters in a larger narrative about what the firm is becoming. Projects are selected and designed because they improve the ability to sense opportunities, make better judgments, and act coherently across functions and time horizons. People experience AI as a collaborator that extends their capability and expands their field of view, reimaging what is possible. The enterprise evolves into a system that learns faster, sees further, and adapts more gracefully than its competitors.
Ultimately, the decisive question is not “how much AI should we deploy?” but “what story will we choose to live with AI?” One story is entangled: tools everywhere, wisdom nowhere. The other is entwined: human judgment and machine intelligence intentionally designed to work together in service of meaningful, previously impossible impact. The underlying technologies may be similar. The difference lies in how the organisation frames its role, designs its collaborations, and commits to a future where insight is genuinely turned into impact.
