Infosys goes model-agnostic as OpenAI builds GSI network to scale Codex

One of seven global partners, Infosys expands its poly-AI strategy as OpenAI leans on GSIs to meet surging enterprise demand for Codex.

Infosys has partnered with OpenAI to accelerate enterprise AI adoption, with a clear focus on taking agentic AI from experimentation to scaled, production-grade deployment.

The collaboration centres on integrating OpenAI’s models and products, including OpenAI Codex, with Infosys Topaz Fabric, Infosys’ agentic AI platform.

The partnership aims to target high-impact enterprise use cases across software engineering, legacy modernisation, DevOps automation and workflow transformation.

By combining Codex with prebuilt agents, workflow automation and enterprise-grade governance, Infosys aims to help organisations redesign development workflows, improve engineering productivity, accelerate delivery cycles and reduce time-to-market.

The broader intent is to move enterprises beyond fragmented pilot projects toward measurable business outcomes at scale.

Infosys chief executive officer, Salil Parekh, said, “Generative and Agentic AI will redefine how enterprises operate and grow. Our collaboration with OpenAI establishes an operating model to unlock AI value at scale – uniting technology, talent, and transformation playbooks so clients can move decisively from pilots to performance, creating competitive advantage.”

“Together, we are not just shaping the future of AI adoption but also enabling our clients to lead it with purpose,” Parekh added.

However, this announcement sits within a larger shift in OpenAI’s enterprise strategy.

Infosys is one of seven global system integrators being onboarded to scale Codex adoption, alongside Accenture, Capgemini, Cognizant, Tata Consultancy Services, PwC and CGI.

This signals OpenAI’s move toward an ecosystem-led model, where GSIs act as the primary execution layer to meet rising enterprise demand for AI deployment.

Infosys builds a model-agnostic AI layer

What stands out strategically is Infosys’ positioning beyond this single partnership. The company is building a “poly-AI” architecture that allows multiple frontier models to coexist within the same enterprise stack.

Its platform is not tied to one vendor, but designed to orchestrate capabilities across providers based on use case, regulatory needs and performance requirements.

This becomes evident in its parallel collaboration with Anthropic, announced earlier, which also runs on the same Topaz platform and focuses on regulated industries including financial services, telecommunications and manufacturing.

Instead of choosing between competing model ecosystems, Infosys is creating an abstraction layer that sits above them.

That approach positions Infosys as a neutral AI orchestrator rather than a model-aligned integrator. For enterprises, it offers flexibility to switch, combine or optimise across models without being locked into a single vendor.

For the broader channel ecosystem, it signals a shift toward services-led differentiation, where value lies in integration, governance and execution rather than the underlying model itself.

At the same time, OpenAI’s push to build a GSI network reflects the scale of demand building around Codex.

As developer adoption accelerates, enterprise use cases are expanding rapidly, requiring partners that can translate model capability into operational deployment.

By leaning on GSIs like Infosys, OpenAI is extending its reach into complex enterprise environments while focusing on advancing its core AI platforms.

Taken together, the collaboration highlights a dual transition underway - OpenAI evolving into an ecosystem-driven platform company, and Infosys positioning itself as a model-agnostic layer enabling enterprises to operationalise AI at scale.