On April 22, 2026, OpenAI announced a partnership with Indian IT services giant Infosys to integrate its tools into Infosys’s Topaz AI platform. The initial focus is software engineering, legacy modernization, and DevOps, with the stated goal of helping enterprise clients move “from experimentation to large-scale deployment.” Financial terms were not disclosed.
This is not a routine licensing item. OpenAI is replaying a classic enterprise-software playbook: where your own sales team can’t reach, distribute through systems integrators and their delivery networks. And for Infosys, it is a move to stay indispensable amid generative-AI anxiety — the fear, across the entire Indian IT sector, that the technology it is being asked to deploy will also automate away the outsourcing work that pays its bills.
What the Deal Covers: Codex Inside Topaz
In practice, OpenAI’s tools — Codex, the coding assistant, is the only product explicitly named in reports — will be embedded into Topaz and paired with Infosys’s delivery capabilities across more than 60 countries, helping clients modernize software development, automate workflows, and deploy AI systems at scale. The initial focus areas, software engineering and legacy modernization, are telling: that is exactly the work Indian IT majors already do in bulk for Fortune 500 clients, which makes Codex a natural insertion point into existing engagements rather than a new budget line customers must be persuaded to open.
Codex’s scale is worth bookmarking: Sam Altman said on X that it has passed 4 million weekly active users.
Codex Labs: The Bigger Rollout That Week
Infosys is one tile in the Codex Labs program announced the same week. Under the program, OpenAI engineers work directly inside client engagements to help deploy its tools; the initial partner list reads like an SI all-star roster — Accenture, Capgemini, CGI, Cognizant, Infosys, PwC, and Tata Consultancy Services.
In other words, this is a deliberate distribution strategy: rather than knocking on every enterprise door, OpenAI turns seven major SIs into its reseller and implementation network. OpenAI had already signed a similar enterprise agreement with HCLTech earlier.
Infosys’s Position: AI Revenue Up, Stock Down
Infosys has strong incentives to lean in. The good news: its AI-related services revenue hit 25 billion rupees (about $267 million) in the December quarter, roughly 5.5% of total revenue — evidence that AI work is already a real revenue line, not slideware. The bad news: its shares are down more than 22% this year, and market coverage of the sector shows HCLTech, Infosys, and TCS all falling sharply together, squeezed by slowing client spending, macroeconomic turmoil, and the fear that generative AI will simply eat traditional outsourcing demand.
Context matters too: Infosys only signed with Anthropic in February 2026 to build enterprise-grade AI agents. Betting on multiple model camps at once is the standard posture for IT services firms in this cycle — none of them wants its fate tied to a single model vendor, and none of the model vendors wants to be locked out of the SI channel either.
Distribution Strategy: Model Vendors Lean on SIs
For OpenAI, the value here is distribution, not technology. The bottleneck in enterprise AI adoption has never been the API; it is integration, data governance, and process rework — exactly the business SIs are in. Codex Labs puts OpenAI engineers on the front line, lowering adoption friction through vendor presence while getting OpenAI tooling wired into the delivery pipelines of hundreds of large enterprises by default.
What It Means for Enterprise Buyers
Three practical effects. First, model selection is becoming an SI-level decision: which model camps your outsourcing partners back will effectively determine which tools end up inside your organization. Second, the move from AI experimentation to large-scale deployment is the defining enterprise theme of 2026 — a line we drew in our opening-of-year outlook — and it is now accelerating through SI networks. Third, for developers, Codex entering the enterprise is not an automatic win: 4 million weekly active users is a strong number, but code governance, security review, and existing toolchains inside each company are where deployment will actually be won or lost.
Sources
- OpenAI teams up with Infosys to bring AI tools to more businesses — TechCrunch
- Scaling Codex to enterprises worldwide — OpenAI
- As AI jitters rattle IT stocks, Infosys partners with Anthropic — TechCrunch
AI-assisted summary compiled from the sources above, reviewed by a human before publishing.