One of OpenAI's largest enterprise deployments ever
Samsung Electronics just announced it's rolling out ChatGPT Enterprise and Codex to all employees in Korea plus the entire global Device eXperience (DX) division. OpenAI is calling this "one of OpenAI's largest enterprise deployments to date."
This isn't a pilot for a single team or a trial for knowledge workers. Samsung plans to use these tools across R&D, manufacturing, marketing, product development, and corporate functions. The scope matters: we're talking about a global manufacturing and technology leader deploying frontier AI across both technical and non-technical workflows at scale.
The relationship goes both ways. Samsung is already working with OpenAI to supply advanced memory semiconductors for next-generation AI infrastructure. Now the partnership expands beyond hardware into workforce transformation.
What Samsung is actually using this for
The deployment covers a surprisingly wide range of use cases. ChatGPT Enterprise will handle knowledge work—searching and analyzing information, drafting documents, developing ideas, interpreting data. Standard productivity stuff, but with enterprise-grade data protection, user management, and security controls baked in.
Codex is the more interesting piece. Yes, it started as a coding tool for software development (writing, reviewing, debugging code). But Samsung is explicitly using it for non-technical teams too.
The example OpenAI gives: employees can "turn ideas into working software, internal tools, websites, and automated workflows." This is Codex as a no-code/low-code platform for internal tooling, not just a GitHub Copilot competitor.
That framing aligns with where Codex usage is headed globally. OpenAI says more than 5 million people now use Codex weekly for "technical and non-technical workflows." And here's a striking data point: Codex weekly active users in Korea grew nearly 800% since February 1, 2026.
Why this deployment is a test case for enterprise AI
Samsung isn't just a big customer—it's a manufacturing company with deep technical complexity. Harrison Kim, General Manager of OpenAI Korea, calls it "particularly significant" because Samsung is treating AI "not as a tool limited to certain teams or functions, but as a core platform for improving how employees around the world work and innovate."
That's the bet: can ChatGPT and Codex handle the operational diversity of a global hardware company? Marketing teams drafting campaigns, engineers debugging firmware, manufacturing teams automating workflows, product managers prototyping internal tools.
The big question is whether ChatGPT Enterprise's security model—data protection, access controls, governance frameworks—holds up at Samsung's scale and complexity. Enterprise AI doesn't fail on capabilities anymore; it fails on trust, compliance, and integration.
If Samsung can thread that needle, it validates the "AI for everyone" enterprise thesis. If edge cases emerge (and they will), we'll learn where the gaps are.
Korea is becoming an enterprise AI testbed
Samsung isn't OpenAI's only major Korean deployment. Seoul National University recently rolled out ChatGPT Edu to all 47,000 students, faculty, and staff at no cost as part of becoming an "AI-native campus." OpenAI also partnered with Kakao to integrate ChatGPT directly into KakaoTalk group chats—Korea's dominant messaging platform.
And the enterprise customer list in Korea is growing fast: LG Electronics, LG Uplus, LG CNS, GS E&C, Samsung SDS, TVING, Krafton, Toss, MUSINSA, Korea Zinc, Nexen Tire, and HanaTour are all using ChatGPT Enterprise, OpenAI APIs, or Codex.
There's a pattern here. Korea has high digital literacy, strong enterprise technology adoption, and a culture that moves fast on infrastructure bets. It's a natural market for testing enterprise AI at scale across diverse industries—electronics, telecom, construction, gaming, fintech, fashion, manufacturing, travel.
If OpenAI can make enterprise AI work in Korea's demanding, security-conscious, operationally complex environment, the playbook probably works everywhere.
What to watch
A few things I'm tracking as this rolls out:
-
Adoption patterns inside Samsung: Which teams actually use this daily versus which treat it as a curiosity? Enterprise software graveyards are full of tools that got deployed but never embedded into workflows.
-
Codex for non-developers: The claim that non-technical employees can "turn ideas into working software" is ambitious. How much does this actually happen, and what does "working software" mean in practice? Internal CRUD apps? Automated spreadsheets? Actual production tools?
-
Security and compliance in practice: ChatGPT Enterprise promises security controls that fit Samsung's governance framework. But Samsung operates across jurisdictions, industries, and regulatory environments. The first time a compliance edge case hits, we'll learn whether the enterprise product is truly flexible or just well-marketed.
-
From deployment to productivity: The hardest part of enterprise AI isn't rolling it out—it's measuring whether it actually improves outcomes. Does Samsung see measurable gains in time-to-market, R&D velocity, or manufacturing efficiency? Or does it become expensive overhead?
OpenAI has landed a marquee enterprise customer with global scale and operational complexity. Now comes the hard part: proving the value holds up in the real world, where AI has to compete with inertia, process debt, and the fact that most people already have workflows that kinda work.
This deployment is a test not just of ChatGPT Enterprise, but of whether frontier AI can actually transform how large, complex organizations operate. I'm cautiously optimistic—but I'm watching the data.