Three years after the spectacular launch of ChatGPT, 2026 could mark a turning point for generative artificial intelligence in the business world. Following an experimental phase characterized by pilot programs and proofs of concept, the time has now come for operational deployment, particularly for so-called agent-based AI.

This shift in scale will have major implications for IT departments, Chief Data Officers (CDOs), and DevOps teams. These stakeholders will need to meet new requirements regarding traceability, accountability, and sustainability for AI models deployed in production.

Agent-based AI is expected to be the next major evolution of generative AI. According to predictions by Gartner, 40% of enterprise applications will incorporate specialized AI agents next year, compared with less than 5% today.

Unlike today's conversational assistants, which are limited to generating content or providing ad hoc assistance to users, these new solutions enable the deployment of truly autonomous agents capable of performing complex tasks, making decisions, and orchestrating end-to-end business workflows.

AI Oversight

This growth is expected to be particularly pronounced in IT operations management. Companies are therefore expected to combine AI agents with AIOps practices, which rely on machine learning models to collect data, detect anomalies, and identify incidents.

From now on, these systems will no longer simply recommend actions: they will intervene directly to solve problems, autonomously.

However, such autonomy requires strict oversight. Companies will need to implement robust governance to ensure that these systems remain accountable, controlled, and aligned with strategic objectives. 

In practical terms, this will involve establishing rigorous standards, conducting tests and audits, and providing detailed documentation for each model (architecture, algorithms, training datasets, performance metrics). Any AI system deployed must be traceable, supervised by human processes, and protected by safeguards.

AI at the Heart of ESG Strategies

The use of AI will also become a central component of ESG (environmental, social, and governance) strategies. Companies will need to redefine their priorities to ensure that their AI systems also meet societal expectations regarding ethics, transparency, and accountability.

This transformation is already leading to the emergence of new roles, such as AI Ethics Officers, as well as the creation of cross-functional committees bringing together legal professionals, engineers, and business leaders. Their shared goal: to make AI not only a driver of performance, but also a source of trust.