Anna Drobakha, Global Digital Business & AI Transformation Director at Groupe SEB.getty​Most organizations have no shortage of people working on AI. They have strategists crafting road maps, technical teams building and deploying, and operators redesigning workflows. The talent is real, and the investment is significant.Yet transformation stalls, not at the start but somewhere in the middle. Initiatives that launched with momentum quietly lose altitude. Pilots that showed promise never scale. The capability that was built sits underused.This isn't a technology problem, and it's rarely a strategy problem. Rather, it's a leadership architecture problem.The disciplines required for AI transformation (strategic direction, capability building and execution) are typically distributed across different people, functions and organizational layers. Each group does its part, but no one owns what happens between the parts. That's where transformation breaks.​I've been thinking about this gap through the lens of a triathlon because it's structurally accurate. A triathlon isn't three separate races. It's one continuous effort across fundamentally different disciplines. What makes it demanding is the requirement to perform across each individual leg without a reset and to manage the transitions without losing momentum.AI transformation works exactly this way across its three disciplines:1. Strategic Clarity: Understanding where AI creates real business value, not in theory but in the specific context of the organization: its operating model, its customers, its constraints.2. Capability Integration: Redesigning how work gets done, how decisions are made, how teams operate with AI embedded rather than added on top.3. Execution Ownership: This is where the work becomes genuinely human. Scaling AI is a mobilization effort that requires leaders who can build a movement inside the organization: clear accountability for outcomes, disciplined governance over what to test, what to stop and what to scale, and visible participation at the top that signals to teams that this isn't another initiative but, rather, how the organization intends to work.Most organizations are capable in at least one of these disciplines. The challenge is holding all three simultaneously and moving fluidly between them as conditions change.​This is where the AI triathlete becomes the right frame. It isn't a job title or a hiring profile. It's a leadership capability that organizations urgently need to develop across the whole of their senior team.The AI triathlete operates across all three disciplines with coherence. They read a strategic signal and translate it into an operational decision this week. They maintain execution momentum without losing sight of whether the work still points in the right direction. They use AI in their own thinking and decision making, which gives them the judgment to lead others through it.What distinguishes them is the capacity to hold all three areas in motion at once and to make confident decisions in the transitions between them.​ Those transitions deserve more attention than they typically receive.In a triathlon, transitions are where races are won or lost. The moments between disciplines (exiting the water, mounting the bike, shifting into the run) are where accumulated effort either holds or fractures. Elite athletes train for transitions as deliberately as they train for each leg.In AI transformation, the equivalent moments are predictable: the gap between a compelling strategy and genuine organizational readiness, the distance between a successful pilot and scaled adoption, the leap from deploying a tool to actually changing how decisions get made. Most transformation efforts stall here because no one was accountable for the handoff.What allows some leaders to navigate those moments while others lose ground is less about expertise and more about how they think. The AI triathlete tends to think in systems. They hold the whole in mind while working on the parts. They see the dependencies between strategy, capability and execution as a live system, where a decision in one area creates conditions (or constraints) in another. This is what allows them to anticipate where energy will dissipate before it happens and redirect it.They aren't reacting to each transition; they're managing the system continuously.​There's one more dimension to this that rarely gets named. In the AI era, endurance doesn't mean stamina. It means treating learning as a permanent operating condition.The tools that worked six months ago may already be underperforming. The prompts that drove results in Q1 need redesigning by Q3. What looked like best practice is already a ceiling for someone moving faster. AI transformation has no stable state to arrive at. The only constant is movement, and the leaders who sustain transformation are the ones who built the discipline of continuous upgrading into how they work and how they expect their teams to operate.​This is also why the AI triathlete is not a role. The instinct to appoint a chief AI officer and consolidate AI accountability in one position is understandable, but it often re-creates the specialist trap at the executive level. AI transformation that lives in a single function, however senior, still doesn't scale. The organizations getting this right are building this capability across the entire C-suite: the CMO who connects AI to revenue, the CFO who models its impact on cost, the CHRO who redesigns how capability gets built. One leader doesn't own AI. Every leader leads with it.​ The human is always in the lead.​The organizations that will pull ahead are the ones developing leaders who can orchestrate across strategy, capability and execution continuously and together. Strength in one discipline wins a lap. The AI triathlete wins the race.​Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?