For many established companies, the biggest competitive threat from artificial intelligence may not be that an existing rival adopts better AI tools. It may be that an entirely new competitor emerges that was designed around AI from day one.

According to Dr Pierre Le Roux, Managing Director of digital consulting firm MOYO, this distinction matters because much of the current corporate AI conversation is still focused on improving the businesses organisations already have.

“Most companies are asking how they can use AI to make existing processes faster, cheaper or more efficient. That is certainly valuable, but an AI-native competitor starts from a completely different position. It asks: if we were building this company today, with AI available across almost every function, how would we design the business?”

The answer, Le Roux says, may look very different from the traditional enterprise. Established organisations carry years, and sometimes decades, of accumulated technology, processes, management structures, reporting lines and approval mechanisms. Many were created for good reasons, but collectively they can make organisations slower and more difficult to change. AI-native businesses have an opportunity to build without much of this legacy.

They can embed intelligent systems directly into workflows, allow data to move continuously between functions and increasingly use AI agents to perform work that previously moved through several teams or organisational layers. “The advantage is not simply automation,” says Le Roux. “The real advantage is that you can potentially build an organisation that senses what is happening, makes decisions and adapts much faster.”

Consider two businesses responding to a change in customer demand. In a traditional enterprise, the information might first be identified by the sales team, analysed by another function, escalated to management, discussed in meetings and eventually translated into changes across marketing, operations or product development. In a more AI-native operating model, intelligent systems could detect the change, analyse its implications, model possible responses and initiate parts of the operational response almost immediately, while humans retain oversight of decisions requiring judgement and accountability.

“One organisation is using AI inside the business it already has. The other has actually been designed to move faster. That difference is going to become increasingly important,” Le Roux says. 

He believes many current discussions around AI transformation remain too narrow. Companies often focus on individual use cases such as customer service, content generation, forecasting, software development or administrative automation. The business case then centres on questions such as how many hours can be saved, which processes can be automated and how much cost can be removed.

Those are worthwhile objectives, but they do not address the bigger strategic issue. “AI has the potential to change the economics of how organisations coordinate work,” says Le Roux. “Functions that historically required large teams may be structured differently. Decisions that used to move through multiple organisational layers can potentially be compressed, and processes that were once rigid can become far more adaptive.”

The risk for established enterprises is therefore that they simply bolt AI onto an operating model that was created for another era. This also challenges the traditional approach to digital transformation. Large organisations have become accustomed to transformation programmes that run for three, four or five years, involving major platform decisions, system migrations and extensive process redesign.

AI is developing too quickly for that approach to be sufficient on its own.

Agentic AI illustrates the problem. The market is already moving from systems that largely respond to prompts towards AI agents capable of planning activities, using tools, interacting with software and completing increasingly complex workflows. “The capability can change significantly in a relatively short period,” Le Roux says. “Businesses therefore need to become much better at continuous adaptation. Transformation cannot be something you do every five years. It increasingly needs to become part of the normal operating model.”

This requires more than simply buying new technology. Organisations need accessible and well-governed data, technology architectures that can accommodate change, clear governance around AI and business processes that can be reconfigured as new capabilities emerge. The rise of AI-native organisations could also have significant implications for the structure of companies themselves.

Le Roux stresses that this does not mean the enterprise of the future will operate without people. Rather, the role of people is likely to change. “Human value increasingly shifts towards judgement, creativity, customer relationships, domain expertise and accountability. At the same time, people should have to spend less time moving information between systems, compiling routine reports or performing repetitive administrative work,” says Le Roux. 

Management structures could evolve too. Many organisational layers developed partly because information needed to be gathered, interpreted and communicated up and down the hierarchy. If intelligent systems can analyse operational information continuously and make it available in real time, some of those structures could change significantly. AI may therefore reshape more than individual jobs. It could reshape the anatomy of the organisation itself.

Established businesses still have formidable advantages. They have trusted brands, existing customers, proprietary data, established distribution, industry expertise and deep knowledge of their markets. But Le Roux cautions that incumbents should not assume these advantages will protect them indefinitely. “The most dangerous future competitor may not be the company you compete with today. It could be a business that does not exist yet, built by a much smaller team, with a very different cost base and an operating model designed around AI from the outset.”

For boards and executive teams, this changes the strategic question. It is no longer enough to ask where AI can improve productivity. Leadership teams should also consider how a new entrant, with no legacy technology, processes or organisational structures, would build their business if it entered the market today. “Would that competitor need the same organisational structure? Would decisions take as long? Would it need the same number of systems, handovers and approvals? And would its cost base look anything like ours?” Le Roux asks.

“These are uncomfortable questions, but they are becoming necessary ones. AI strategy is rapidly becoming operating-model strategy. The businesses that understand that early will not simply use AI to make the organisation they already have more efficient. They will use it to rethink what the organisation should become,” concludes Le Roux.