GeethanTechGeethanTechTech, decoded daily.
LatestAISoftware & PlatformsCybersecurityStartups & VentureTech Business & Markets
All posts
AI5 months ago

Samsung Rolls Out ChatGPT Enterprise and Codex to Staff

Samsung is making ChatGPT Enterprise and Codex available to employees in Korea and DX staff worldwide, signaling a wider shift from AI pilots to workplace platforms.

Samsung Rolls Out ChatGPT Enterprise and Codex to Staff

Samsung Electronics is making ChatGPT Enterprise and Codex available to a much wider employee base, turning its OpenAI partnership from an infrastructure story into a workplace AI rollout. OpenAI announced the deployment on June 21, 2026, and The Korea Times reported on June 22, 2026 that the deal makes Samsung one of OpenAI's major ChatGPT Enterprise customers.

Under the agreement, ChatGPT Enterprise and Codex will be available to all Samsung Electronics employees in Korea and to all employees worldwide in the company's Device eXperience division. OpenAI describes it as one of its largest enterprise deployments to date, which makes the rollout a useful signal for how large technology companies are beginning to operationalize AI tools at scale.

The important detail is that Samsung is not limiting the tools to a small engineering pilot. The company plans to use ChatGPT and Codex across software development, marketing, product development, manufacturing, corporate functions, and other areas of the business. That makes the deployment less about one chatbot license and more about treating AI as a shared work layer across technical and non-technical teams.

For everyday knowledge work, ChatGPT Enterprise is meant to support tasks such as searching and analyzing information, drafting documents, developing ideas, and interpreting data. OpenAI says the enterprise version includes data protection, access management, and security controls, which matters because large companies cannot usually bring broad AI tools into internal workflows without governance around users, data, and permissions.

Codex is the more developer-facing part of the rollout, but OpenAI is positioning it as broader than code generation. The tool can help write, review, and debug code, while also helping employees turn ideas into internal tools, websites, and automated workflows. That is why this deal matters for GeethanTech readers: the same agentic coding tools that started inside engineering teams are now being framed as business automation tools.

OpenAI also says Codex now has more than 5 million weekly users across technical and non-technical workflows, and that weekly active users in Korea have grown nearly 800 percent since February 1, 2026. Those figures do not prove the Samsung rollout will work perfectly, but they explain why enterprise software buyers are treating AI agents as a platform category rather than a narrow productivity add-on.

The partnership also extends an earlier OpenAI-Samsung relationship around AI infrastructure. Samsung has been working with OpenAI on advanced memory semiconductors for next-generation AI systems. With ChatGPT Enterprise and Codex now moving into Samsung's employee workflows, the connection between the two companies spans both the hardware required to run AI and the software tools employees use to apply it.

The bigger takeaway is that enterprise AI adoption is becoming more practical and more demanding at the same time. Deploying ChatGPT and Codex to a company as large and complex as Samsung creates room for faster analysis, software work, and internal automation, but it also raises the bar for training, workflow design, security policy, and measuring whether AI tools actually improve how teams work.

Continue with this topicAI
Browse categoryarrow_forward
Continue reading

Related posts

Three linked icons labeled Commit, Prove and Verify, ending in a green check mark.
AIabout 12 hours agoBy megan

Anthropic’s FlashProve prototype checks AI output one token at a time

Anthropic tested cryptographic proofs for a quantized Llama 4 Maverick configuration, allowing a CPU to check a token against committed model weights without seeing them. Its 1,024-token context and high proving cost leave routine API use unproven.

Read article
GDPx chart showing Claude Opus 5.5 at a 12.0% composite score on a 0 to 100 percent scale.
AI3 days agoBy megan

GDPx gives its top AI model a 12.0% composite score

SemiAnalysis’s new benchmark tests agents on 206 tasks across 11 domains. Its headline score combines different grading methods, so it is not the share of tasks solved.

Read article
GeethanTech
GeethanTech
Tech, decoded daily.
Read
Latest postsAISoftware & PlatformsCybersecurityStartups & VentureTech Business & MarketsRSS feed
Connect
mailEmaillanguageWebsiteinLinkedInIGInstagramfFacebook@Xsmart_displayYouTube
From the publisher

Read all our publications: GeethanPost | GeethanTech

A chapter in the ElegantHumanity story

Powered byElegantArc