A report published by Sopra Steria Next predicted that the global AI market could grow from $540 billion in 2023 to $1,270 billion five years later. This rapid growth should push companies to accelerate their transformation.
It is an uncomfortable injunction that has recently entered the executive committees and boards of companies: with the arrival of generative AI and the multiplication of AI's potential, only companies capable of deploying it operationally would be successful. While this prediction is certainly radical, it remains – partially – true. After several years of testing AI on different projects, many companies find themselves stuck at the industrialization stage, unable to move from in vitro testing to in vivo deployment. The finding is unequivocal: today, only one in seven AI algorithms developed in companies reaches the deployment stage. An 85% failure rate that illustrates the difficulty organizations face in industrializing their AI initiatives. Yet, the stakes are high: the global AI market, valued at $540 billion in 2023, is expected to more than double by 2028 to reach $1,270 billion.
A strong need for a new way of understanding AI
"Leaders should not see artificial intelligence as a technological issue. The key today is to approach it as an business one." explains Fabrice Asvazadourian, CEO of Sopra Steria Next.
The problem leaders face today is quite simple: the market lacks tools and methodologies to economically approach artificial intelligence. For two decades, AI providers have been talking to IT departments in a language rooted in the technological universe (Machine learning, deep learning, or even LLM or RAG more recently). But knowing how these technologies work does not help to deploy them.
"We see it every day, CEOs are perfectly aware of the immense economic challenge that artificial intelligence represents at a macro level, and at the same time, they fail to make it a profitable technology in their own organization," says Bruno Maillot, Director of the AI for Business practice at the firm.
To offer these leaders a tailor-made approach to AI, Sopra Steria Next has developed an innovative methodology. The consulting firm recommends addressing the issue by focusing on uses rather than technology.
Focusing on Uses
This method revolves around four uses that allow AI to be sequenced based on what is done with it. These main types of AI can be classified into four main groups.
- AI for Machines focuses on optimizing equipment, factories, and supply chains – it mainly concerns transport, security, industry, and real estate.
- AI for Processes aims to streamline management activities and significantly impacts financial services, public services, support functions, and commerce.
- AI for humans covers next-generation tools, knowledge management, and various virtual assistants – it will mainly develop within financial services, telecommunications, and healthcare.
- AI for software development encompasses everything related to the complete cycle of IT development in companies and, of course, affects companies and services involved in development or software publishing.
"We want to offer leaders a strategic compass in the complex and multifaceted world of AI. This market structuring allows optimizing the company's investments and ensuring that there are no blind spots in their exploration of AI's various potentials," details Fabrice Asvazadourian.
Taking these factors into account, the consulting firm estimates that:
- The AI for Machines market will grow by 13% per year and should reach $330 billion by 2028, representing 26% of the global AI market.
- The AI for processes one is growing by 18% per year, reaching $390 billion by 2028, or 31% of the AI market.
- AI for humans will grow from $130 billion to $380 billion in five years, occupying 30% of the AI market, marking the highest growth in volume.
- AI for software will triple, reaching $170 billion with an annual growth of 25%.
Knowing where to invest
The analysis of uses constitutes a strategic lever to optimize investments in artificial intelligence. It also offers companies a valuable tool for sectoral comparison. In financial services, for example, the firm recommends focusing on AI for humans and process automation. This analysis allows actors to evaluate their positioning and adjust their strategy accordingly.
Based on the insights from this study, Sopra Steria Next shares its recommendations on the deployment of AI in companies. The firm's experts notably recommend concentrating the bulk of investments (80%) on use cases that have already proven themselves in the sector, intelligently combining predictive and generative AI. They also remind us that the human factor is crucial: it is not only about attracting and training specialized AI technical talents but also about supporting all employees towards a smooth adoption of these new tools.