This is the Contrarian view: AI Is Overhyped At Best, Incremental is the Best
Contrarians often argue that AI is impressive but does not change the way businesses compete. AI, from this perspective, is just another tool for productivity, like spreadsheets, ERP or cloud computing. It is useful but not transformational.
This view is supported by those who believe that AI will quickly lose its advantages. As long as every company is able to access the same models, agents, tooling and tools, then AI will not be an advantage. The margins become normalized, the differentiation is lost, and success becomes a function of brand strength and execution quality.
You can also find out how many AI projects are not performing as expected. Agents need supervision and models hallucinate. Data quality issues also reduce promised returns. AI in this context is used to speed up or reduce the pressure on headcounts without changing business models.
It is a view that has a historical foundation and is grounded in reality. This is because many past technologies that promised revolution delivered instead optimization. It is not the case that this argument is wrong all of the time, but rather that it presumes that organisations remain unchanged. AI is a sluggish technology when it’s forced to function within the constraints of legacy incentives, workflows and organizational charts.
Provocative AI: Views in 2026
AI will hollow out traditional organizations, according to the more aggressive view
AI won’t just improve businesses, is a more aggressive and uneasy position. This will reveal how many corporate structures are designed primarily for humans to be coordinated, rather than creating value.
In this view, middle management layers, coordination roles and entire departments can be viewed as optimization artifacts from a pre AI world. AI agents can execute and monitor tasks, eliminating the need for all of these layers. It is left with a small team of high-level decision makers, while AI handles the majority of operational execution.
Companies that are still clinging to their traditional structures, with heavy headcounts, will be outcompeted in the future by AI-native companies, which have lower costs of operation and quicker decision-making loops. Not only is the disruption technological, but also organizational. The company itself shrinks, flattens, and becomes volatile.
It is implied that AI’s advantage does not actually come down to productivity. The question is who will be willing to demolish parts of their organization which no longer make any sense even though it may cause cultural and political pain.
What More? Pessimistic ViewAI Does Not Matter As Much as Said
The pessimistic viewpoint is that AI won’t provide any meaningful advantage to most companies. According to this argument AI capabilities are likely to commoditize quickly, while regulation and risk-aversion slow the deployment of AI.
AI will be a technology that all firms have, but very few trust. The decision-making process is still human, because the accountability of decisions cannot be automated. AI’s advisory role is more likely to be a result of errors, regulatory oversight, and concerns about bias. There are productivity gains, but these are small and unequally distributed.
AI will not be transforming industries in the future but rather quietly integrating into software platforms. In this future, the winners will not be those who have superior AI systems but rather those that excel in strategy, price power and customer relations. AI will become background infrastructure, not a disruptive force.
This view does not have the danger that it’s implausible. The danger is not that it is implausible, but that companies that adopt AI too early could miss out on structural changes that are still possible. In the event that AI proves to be transformational, those who adopt it late will find themselves unable to catch up by simply buying similar tools.
