NVIDIA consolidates its agentic stack with the acquisition of ILLUMEX

Nvidia has just finalized the acquisition of Israeli startup Illumex for around $60 million, or nearly 51 million euros. Illumex works on the semantic structuring of enterprise data to make AI systems interpretable, reliable and actionable in complex environments. This acquisition highlights an evolution of Nvidia, which strengthens its integrated stack, designed for agentic AI in business.

A structural weakness of AI in business

Companies experimenting with generative AI encounter a recurring obstacle: the distance between natural language and organizational reality. Models include texts, when organizations operate with repositories, business rules, nomenclatures and regulatory constraints.

Illumex has developed a platform called Generative Semantic Fabric (GSF), which transforms a company’s information corpus into structured business ontologies. These ontologies formalize the entities, their relationships, their exact definitions, which then serve as a basis for AI agents to interpret queries, produce analyzes or execute workflows.

In a financial, industrial or health environment, an approximation can become an operational risk. By attacking the semantic layer, Illumex tends to reduce these hallucinations.

🚨 SMARTJOBS

  • ECOLE POLYTECHNIQUE – Director/Deputy Director of International Relations (F/M)
  • CLAROTY — Sales Development Representative
  • CURE51 — Data Scientist (Internship)
  • FRACTTAL — Account manager (France)
  • BRICKSAI — Founding Growth Manager

👉 Find all our offers on the DECODE MEDIA Jobboard

đź“© Are you recruiting and want to strengthen your employer brand? Discover our partner offers

Secure adoption rather than seduce by performance

The competition around fundamental models has long been based on benchmarks and comparative performances; the companies’ reading grid focuses more on the reliability and traceability of AI. It is in this sense that Nvidia is no longer limited to providing the computing power that trains the models but seeks to facilitate their integration into structured and regulated environments.

This trend is found among hyperscalers, who develop their own orchestration and governance layers, but also SaaS publishers who integrate specialized agents. Controlling the semantic layer means reducing dependence on third-party solutions and strengthening the consistency of the offer.