As AI gradually is embedded in their decision-making processes, companies must find the right balance between technological promise, operational constraints, risk management and value creation. Data quality, governance, AI agents, human expertise: explore key innovations and strategies that will fundamentally transform your organisation in the years to come.
Turning initial uses into long-term added value for the organisation
On everyone’s lips and everywhere in our daily routines – from the simplest to the most complex – artificial intelligence has (definitely) moved beyond the experimental phase. For organisations, the question is therefore no longer whether they will adopt AI, but how they can scale it up reliably and sustainably. However, there remains a significant gap between the excitement generated by its potential and the reality of its impact within companies: whilst AI features heavily in companies’ stated strategies, it is, in contrast, much less (or poorly) integrated into decision-making processes.
Coface's latest edition of the Data & AI Barometer 2026, conducted with Les Echos Etudes amongst 900 decision-makers, reveals very different levels of maturity depending on company size: the larger the company, the more integral AI is to the decision-making process. Today, one-third of medium-sized and large enterprises consider themselves to be well advanced in their use of AI, compared with one in ten SMEs. And whilst two-thirds of decision-makers trust their data, one-third consider it to be poorly aligned with business needs and tools for managing risks and informing decisions.
Data quality is, more than ever, the key factor in the successful industrialisation of AI within a company, enabling it to move faster and achieve efficiency gains.
The resources, skills and capacity to structure Data & AI projects seem to favour large organisations. This carries the risk that the gap will become increasingly difficult for small businesses to bridge as AI applications become more structured,
explains Nesrin Gonin, Business Information Director - Coface Western Europe & Africa.
No data governance, no performing AI
For any company, regardless of its size or sector, the scalability and added value of AI depend heavily on the available data’s quality: however powerful it may be, an AI model can never sustainably compensate for information that is obsolete, fragmented or poorly defined. Yet 60% of companies surveyed by Coface still use AI without a formal framework. At this pivotal moment for organisations, and as the use of AI becomes more widespread, the issue of how to manage it is becoming urgent. Yet, for the great majority of companies, data governance has yet to be implemented: this is precisely where limitations are quickly reached, which most often results in disappointing outcomes for the company and its business teams.
Data governance depends on reliable data, updated in real time, centred on business needs, and the creation of a common semantic layer: that is to say, sharing the same definitions for each indicator and the same context of use. In practice, the same data can in fact represent several different realities depending on whether it is used by Risk, Finance, Procurement or Sales teams.
AI acts as a catalyst: it amplifies both the quality of the data and its weaknesses. As it becomes more widespread throughout the organisation, establishing clear rules on data, usage and responsibilities is becoming essential.
Data governance is no longer a peripheral issue: it is a prerequisite for creating sustainable value with AI,
says Samia Boujatioui, Group Head of Data Management – Coface Data Office.
Business alignment, technological innovations: making faster yet better decisions
The next major breakthrough in the full adoption of AI within organisations could come from AI agents connected to corporate data. Their potential is tangible:
directly query a customer portfolio,
identify companies whose risk profile has deteriorated,
explain a trend or highlight comparable business opportunities.
This development meets a very pragmatic need. Nearly one in two decision-makers wants to obtain clear and immediately actionable insights through AI, even if the aim is not to blindly delegate decision-making to a machine. Finance departments, for example, want to strengthen fraud detection and better anticipate customer risks, whilst sales teams are increasingly looking to identify new markets.
In the future, the advantage will not necessarily lie with the company that collects the largest volume of data, but with the one that can align the real needs of business teams with the power and security of technology and ensure harmonised governance at every level of the organisation.
What will set apart a company that has successfully completed its Data & AI transformation is its ability to turn its data into concrete, faster and more effective decisions to generate a genuine business advantage.
states Nesrin Gonin.
Stay ahead of the game and make decisions with complete confidence
> What if managing risk were a way of staying open to opportunities? Learn more about the global study and access the Open Advantage roadmap and its five strategies for turning risk into a driver of growth.
