Samsung reverses its ChatGPT ban with a global Codex rollout
Three years after barring generative AI over a source-code leak, Samsung is handing ChatGPT Enterprise and Codex to its global workforce. India is watching closely.
The News
Samsung Electronics is rolling out OpenAI's ChatGPT Enterprise and the Codex coding agent to its staff, in what OpenAI describes as one of its largest enterprise deployments to date. The rollout reaches employees across Korea and the company's global Device eXperience (DX) division, the unit that builds Samsung's phones, televisions and home appliances. Announced in mid-June 2026, the tools are being aimed at software development, marketing, product development, manufacturing, R&D and corporate functions.
The decision closes a remarkable loop. In March 2023 Samsung banned generative AI internally after engineers pasted sensitive source code and meeting notes into ChatGPT, leaking proprietary material. A little over three years later, the same firm is handing those tools to its workforce at scale, with full global training targeted by the end of 2026.
Samsung is not betting on a single vendor. Alongside OpenAI's products, it is also deploying Google's Gemini and Anthropic's Claude. Distribution runs through Samsung SDS, the group's IT services arm, which agreed a reseller partnership with OpenAI in late 2025.
Why It Matters
The story is less about one more enterprise licence and more about how fast corporate caution on AI has flipped. When ChatGPT first reached workplaces in early 2023, a wave of large employers including JPMorgan, Verizon and Amazon restricted or blocked it over data-leak fears. Samsung's own ban was among the most cited examples of that anxiety. Its reversal is a signal that the leak-then-embrace arc has become the default corporate journey, not the exception.
What changed is the packaging. Enterprise tiers now promise that company prompts stay out of public training data, and coding agents such as Codex have moved from novelty to measurable output. For a hardware giant whose competitiveness rests on chip design, firmware and manufacturing software, putting an autonomous coding assistant in front of engineers is a direct bet on development speed rather than a marketing experiment.
The multi-vendor approach matters too. By running OpenAI, Google and Anthropic models side by side, Samsung avoids lock-in and keeps pricing leverage, a posture more buyers are likely to copy as the model market commoditises.
Indian Angle
For India, this is not a distant Korean story. Samsung operates some of its largest research centres outside Korea in Bengaluru, Noida and Delhi, with tens of thousands of engineers working on mobile, networks and appliance software. A global DX rollout means those teams sit squarely inside the deployment, making India one of the bigger real-world tests of whether Codex actually lifts engineering throughput.
It also sharpens the question facing Indian IT services majors such as TCS, Infosys and Wipro, whose business depends on the same software-development hours that coding agents now compress. Samsung internalising AI tooling at scale is a preview of the productivity pressure Indian outsourcing must answer, ideally by reselling and integrating these systems rather than being displaced by them.
There is a governance lesson as well. Samsung's original ban was a data-protection failure, and India's Digital Personal Data Protection Act gives local enterprises a sharper reason to insist on enterprise-grade controls before letting staff loose on public chatbots. Domestic model builders such as Sarvam and Krutrim, meanwhile, will note that even a deeply Korean conglomerate is hedging across three foreign vendors, a reminder of how much room a credible Indian enterprise alternative still has to claim.
FAQ
When does the rollout take effect?
Samsung began the deployment in mid-June 2026, starting with staff in Korea and its global DX division. Full global workforce training is targeted by the end of 2026, so access should widen across regional offices through the second half of the year.
How does this compare to Samsung's 2023 stance?
It is a near-complete reversal. In March 2023 Samsung banned generative AI after a source-code leak; in 2026 it is distributing the same class of tools through Samsung SDS, this time wrapped in enterprise controls meant to keep company data private.
What does it mean for Indian engineers?
Samsung's Bengaluru, Noida and Delhi R&D teams fall inside the global DX rollout, so thousands of Indian engineers gain Codex and ChatGPT Enterprise access. It also intensifies the productivity debate already running through India's IT services sector.
Where can I read the original announcement?
OpenAI published the news on its company blog, and the source link appears in the attribution paragraph below.
This story was reported by OpenAI. Read the full original coverage at OpenAI.