Artificial intelligence enters the enterprise: the real challenge is to grow it in processes

The main obstacle, in fact, is not the algorithm. It is the ability of the enterprise to build the conditions for AI to really work: reliable data, integrated processes, clear governance and an architecture capable of evolving over time.

I say this because a study we conducted in Italia interviewing 575 medium to large companies reveals a gap in confidence about the quality of data that is the basis for AI. A significant proportion of the managers interviewed said they lacked confidence in their organisation's ability to integrate data across business functions (34 per cent), lacked quality data (30 per cent) and struggled to overcome fragmented organisational silos (27 per cent).

This shows that the real obstacle is not the algorithm, but data maturity and organisational integration.

The second point is that artificial intelligence produces value when it gets to the heart of processes: from supply chain to finance, from human resources to customer management.

This is why I believe the market needs a pragmatic approach. Pragmatic means starting with the processes where the value is already present, integrating AI directly into business processes, without building separate structures that are expensive and difficult to maintain.

In other words, there is no need to add an extra layer, there is a need to make artificial intelligence live within the processes, where the business context already resides.

This approach radically changes the payback time: no longer months or years for project implementation, but weeks, as well as allowing for progressive, sustainable and scalable adoption.

The real evolution will be the emergence of an increasingly autonomous company, in which specialised intelligent agents collaborate with each other through different processes, supporting people in higher-value activities and increasingly taking over operational and decision-making tasks.

This is not about replacing human labour, but about increasing its potential. The companies that know how to govern this transition with executive discipline, data quality and a clear architectural vision will be the ones that turn AI into a real competitive advantage.

Artificial intelligence is not lacking in companies today. What is often lacking is the ability to integrate it into the business environment, processes and decisions. It is on this terrain that the competitiveness of companies in the coming years will be measured.

(*) Managing Director of SAP Italia