AI is dismantling the foundations most senior professionals built their positions on. One source of advantage remains structurally out of its reach.
Two groups of senior professionals are feeling the same shift from different angles.
The first built their power base on expertise others did not have. Years of accumulated specialist knowledge, pattern recognition refined across thousands of decisions, a depth of domain understanding that took decades to develop and could not be easily replicated. That expertise was the asset. It justified the authority, the fees, the seat at the table, and the power that came with being the person who knew what others did not.
The second built their power base on market insight and analytical capability. Access to better information, faster analysis, more rigorous frameworks applied with more sophistication than competitors could match. The ability to see the market more clearly and move on that clarity before others could. Executives who commanded rooms because their judgement was backed by analysis nobody else had access to.
Both power bases are facing the same structural disruption. And both groups are asking the same question, even if they would phrase it differently. If AI can produce in days what took years to develop, where does my advantage now come from.
What AI has dismantled
Large language models are absorbing and reproducing specialist knowledge at extraordinary speed. The depth of domain mastery that once took years to accumulate, and that competitors could not easily replicate, is now accessible at near zero cost to anyone with sufficient general capability and the intelligence to apply it. Not with the nuanced judgement that genuine mastery produces. But sufficiently to erode the information asymmetry that once made specialist knowledge a defensible competitive position.
The analytical advantage has collapsed in the same way. The frameworks, the market analysis, the strategic options that once justified significant investment in research and advisory capability are now within reach of a capable individual who combines broad experience with AI mastery. That individual does not need the decades of specialist accumulation the established player assumed was their moat. They arrive lighter, faster, and unconstrained by the institutional commitments that slow the incumbent’s response.
The competitor who has replaced your analytical edge may not look like a competitor yet. But the barrier that kept them out is crumbling.
The established player faces something the new entrant does not. Every significant response to this threat requires a decision that challenges what the incumbent fundamentally is, what it has built, and what it has told its market, its people, and itself that it stands for. That is a Galileo Dilemma of the most difficult kind. The new entrant arrives without that baggage. They have no legacy position to protect and no system identity to defend. They can move with a freedom the incumbent cannot match without first navigating a structural condition most organisations do not yet have a name for.
See also: The Galileo Dilemma.
This is not a gradual erosion. It is a structural collapse of the information asymmetry that underpinned competitive advantage for most of the last half century. And it is exposing something that was always present but rarely examined with sufficient clarity because it did not need to be. There were easier sources of advantage available.
What is being exposed
Original thought has always been a source of competitive advantage. What AI has done is remove the sources that obscured it and reveal it as the one that cannot be replicated by any current AI architecture.
Original thought is the capacity to see what the system cannot see. To identify the question the established frameworks cannot ask. To hold an insight against the weight of institutional resistance long enough for it to matter. To move before the consensus has formed and the window has closed.
Large language models cannot produce this. Not because they lack processing power or analytical sophistication. Because original thought by definition departs from the weight of existing published thought that LLMs are trained on. They are extraordinarily capable consensus machines. The insight that genuinely challenges the consensus is precisely what their architecture cannot generate.
This is not a temporary limitation awaiting a technical fix. It is structural. And it means that as AI dismantles every advantage built on information asymmetry, it is simultaneously exposing original thought as the source of advantage it could never touch.
Why it is so rare
If original thought is where advantage now concentrates, the question worth sitting with is why so few senior professionals and organisations are able to access it consistently.
The answer is structural. Every system tends to produce ideas compatible with its own existing architecture. It evaluates thinking by the standards it already holds. It rewards what fits and resists what threatens. Original thought, the kind that would genuinely challenge the system’s existing architecture, rarely emerges from within that system and almost never without cost.
The senior professional who has spent two decades building expertise and authority within a system has been shaped by that system as much as they have shaped it. The thinking that served them well is also the thinking the system trained them to produce. Departing from it feels risky. The system around them will resist it even when they manage to depart.
This is not a personal failing. It is a structural condition that operates in every system at every level. It has a name. And understanding it is where the response to this shift begins.
The colleagues and competitors still optimising for information advantage are running faster toward a destination that is no longer there. The organisations still commissioning the same analysis from the same sources are paying for consensus at precisely the moment when consensus has become the least defensible position available.
The professional who develops the capacity for original thought is not competing on the same terms as everyone else. They are operating in territory that AI cannot colonise, that cannot be purchased from a consultancy, and that cannot be replicated by a competitor the following Tuesday.
That territory is not comfortable. It requires a different relationship with the systems you operate within and a different understanding of what your power base is actually built on. But it is where genuine advantage now lives. And the window for moving there deliberately, before the next wave of disruption closes it, is open now.
See also: Repositioning Your Power Base
Colin Gautrey, May 2026
Colin Gautrey works privately with senior leaders who realise their next move requires original thought.