Artificial Intelligence in 2025

The report “AI in 2025: current initiatives and challenges in large enterprises”, produced by Wavestone in collaboration with French Tech Grand Paris, provides a comprehensive overview of the adoption of artificial intelligence in large organizations in 2025. Here’s a detailed analysis of the key findings.

AI Governance: Between Compliance and Performance

The rise ofgenerative AI (GenAI) has forced companies to rethink their approach to governance. According to a Gartner study, over 80% of organizations have had to adapt their governance frameworks to integrate the specificities of generative AI. This evolution is accompanied by a transition from a purely technological approach to a more strategic vision, aligned with business objectives.

Integrating AI with existing practices in data governance, cybersecurity and human resources is becoming crucial. The European AI regulation is playing a leading role in defining global standards, despite some criticism of its complexity.

AI in Action: From Use Cases to Adoption

Traditional AI technologies, such as Machine Learning and Computer Vision, are now well established in large companies. According to a McKinsey study, over 60% of organizations use these technologies on a regular basis. The emergence of generative AI marks a new phase, promising significant advances in productivity and personalization. Companies are progressively moving from proof-of-concept to large-scale deployment, taking care to align business and technological objectives.

Technical aspects: Traditional vs.

Companies today need to strike the right balance between traditional and generative AI. According to an analysis by Deloitte, the combination of the two approaches often delivers the best results. Centralized platforms connect enterprise data, while separate pipelines address specific needs. MLOps and LLMOps are emerging as essential methodologies for streamlining the development of AI systems.

Security and trust

AI intensifies existing risks and introduces new threats. According to an IBM report, AI-related incidents increased by 40% in 2024. Emerging frameworks, including theEU AI Act and the NIST AI RMF, are guiding organizations on cybersecurity and privacy. Incident management involving AI models is becoming increasingly complex, driving many organizations to outsource this function.

Ethical and environmental responsibility

The democratization of AI raises major ethical questions, particularly with generative AI. According to a study by MIT Technology Review, the environmental impact of generative AI is significant, particularly in terms of energy and water consumption. Companies are trying to promote frugal practices and green engineering strategies, such as the use of Small Language Models (SLM). The lack of supplier transparency and standardized metrics remains a problem.

Workforce transformation

Traditional AI has already revolutionized business processes over the past decade. According to a World Economic Forum study, generative AI will impact the workforce more broadly. Companies are deploying “Secure GPT” systems and developing organization-wide AI solutions. HR must anticipate change, redefine objectives and manage the impact on the workforce.

The Talent Race

AI is generating global competition for talent, with significant differences in salaries and availability across regions. According to LinkedIn, demand for AI professionals has risen by 74% by 2024. The talent shortage is due to the rapid pace of AI innovation outpacing that of training programs. Up-skilling and retraining strategies are becoming essential.

Strategic Autonomy in the AI Era

Global AI competition is dominated by the USA and China. According to Stanford’s AI Index report, Europe is significantly lagging behind, risking technological dependence. Companies must pursue their strategic autonomy, aligning public and private interests. A clear industrial policy on AI is needed, favoring local ecosystems in key sectors.

Future prospects

The report highlights several important future trends:

– The emergence ofmultimodal and multi-agent AI
– The development of more frugal AI models
– The strengthening of global AI governance
– The growing importance of ethics and responsibility

Recommendations for companies

Organizations should:

1. Develop a clear AI strategy aligned with their objectives
2. Invest in training and talent development
3. Strengthen their governance and security
4. Promoting an ethical and responsible approach
5. Maintain technological autonomy

Conclusion

The year 2025 marks a turning point in the adoption of AI by large companies. The challenges are many: governance, security, ethics, talent, but the opportunities are considerable. Success will depend on organizations’ ability to navigate this complex environment while maintaining a balance between innovation and responsibility.

Additional source: the OECD Economic Outlook on AI offers further insight into the macroeconomic impacts of this transformation.