This new study, entitled "Generative AI – From Exploration to Impact", explores the dynamics of the generative artificial intelligence market, the challenges to be met, and the opportunities to maximize the value of generative AI.
In brief:
- 2024 was a year of technological acceleration for generative AI with unprecedented investment levels and the emergence of more advanced and varied solutions.
- Some disappointment among companies, with only 1 in 5 managing to deploy a first use case at scale, despite the proliferation of prototypes ("POCs").
- A confirmation of the generative AI market projection to $100 billions by 2028.
The generative AI market more than doubled in 2024, reaching between $20 and $25 billion in revenues and confirming its projections for the generative AI market, which is expected to reach at least $100 billion by 2028 in the central scenario.
But many obstacles remain within companies
Despite this technological boom, many obstacles remain within companies. Indeed, only 22% of large companies managed to deploy at least one generative AI use case at scale in 2024.
Companies face a paradox: they are fully aware of the potential of generative AI, but still find it difficult to industrialize it and extract tangible value from it. These blockages are not technological but primarily organizational and operational.
From exploration to impact
To support its clients in this transformation, Sopra Steria Next has identified four strategic axes with the aim of accelerating the adoption of generative AI:
- Focus on tangible impact on the P&L – Successfully deploy the most mature generative AI solutions at scale and win the battle of daily use by employees.
- Explore the possibilities of agentive AI – Move from "Text-to-Text" to "Speak-to-Action," and move towards an integrated and personalized multi-task approach: "Smart Lean".
- Learn to combine different generative AI models – Know how to use multiple generative AI models and make them work together to optimize the Cost / Performance / Speed / ESG footprint quadrant.
- Ensure ethical and responsible AI deployment at scale – Possibility of using synthetic data to improve and simplify data management, and of course integrate AI regulation (AI Act).
Click here to discover a summary of the main figures and lessons from the study.
Or complete the form below to ask the full version.