OpenAI June 2026: Samsung Deal, GPT-5.5-Cyber, IPO

OpenAI closed June 2026 with three significant moves: a global enterprise deployment with Samsung Electronics, the full release of GPT-5.5-Cyber for security professionals, and continued structural steps toward a potential IPO — all while CEO Sam Altman publicly expressed ambivalence about leading a public company.

Samsung Electronics Deploys ChatGPT and Codex Globally

Samsung Electronics began rolling out ChatGPT Enterprise and Codex to employees worldwide on June 21, 2026, in what OpenAI described as one of its largest enterprise deployments to date. The agreement covers all Samsung Electronics employees in Korea and all employees in its Device eXperience (DX) division globally.

According to the OpenAI blog announcement, Samsung plans to use both tools across software development, marketing, product development, and manufacturing. ChatGPT Enterprise handles knowledge-based tasks — document drafting, data interpretation, information retrieval — while Codex targets software workflows including writing, reviewing, and debugging code.

OpenAI emphasized that ChatGPT Enterprise includes enterprise-grade data protection, user and access management, and security controls, enabling Samsung to operate within its existing governance framework. The Samsung deal follows a pattern of OpenAI scaling enterprise contracts with large multinationals, with the company citing it as one of the largest agreements in its history.

GPT-5.5-Cyber Launches With 85.6% on CyberGym Benchmark

OpenAI released the full version of GPT-5.5-Cyber as part of an expanded Daybreak cybersecurity initiative, positioning it as the company’s most capable model for authorized security work. Access remains restricted to verified defenders.

According to SecurityWeek’s coverage, GPT-5.5-Cyber scored 85.6% on the CyberGym benchmark — which tests whether an agent can reproduce known vulnerabilities — compared to 81.8% for the standard GPT-5.5. The model can sustain analysis across large codebases, assess whether vulnerable code is reachable, and carry work through to patch development and testing.

OpenAI’s broader argument with Daybreak is that AI has accelerated vulnerability discovery to the point where patching, not finding bugs, is now the primary bottleneck for defenders. To address this, the company also released an updated Codex Security plugin that scans entire codebases, traces attack paths, constructs threat models, validates findings, generates patches, and exports results via SARIF files and CodeQL queries.

Since a research preview launched in March 2026, Codex Security has processed more than 30 million commits across over 30,000 repositories, with human reviewers confirming more than 70,000 fixes and an additional 500,000 findings resolved automatically, according to SecurityWeek. OpenAI also unveiled Patch the Planet, a new initiative co-founded with Trail of Bits and developed with HackerOne.

Omio Cuts Development Time 80% Using OpenAI APIs

Travel platform Omio reported measurable efficiency gains from its OpenAI integration in a June 23, 2026 case study published on the OpenAI blog. The company, which connects travelers across more than 3,000 transportation providers in 47 countries, said new product builds that previously required multiple developers over three months now take a single developer one month — representing roughly 80% less development effort.

Omio CTO Tomas Vocetka described the company’s shift toward conversational travel interfaces, where users describe a desired journey and receive personalized, bookable itineraries rather than navigating multiple booking sites. The platform uses ChatGPT, Codex, and OpenAI’s API.

The Omio deployment illustrates how OpenAI’s enterprise strategy extends beyond large corporations into mid-market technology companies, with the travel sector as a growing vertical.

Sam Altman on AGI, IPOs, and OpenAI’s Direction

Sam Altman made two notable public statements in June 2026 that reflect the competing pressures on OpenAI as an organization.

On AI timelines, Fortune reported that Altman predicted AI will surpass human intelligence by 2030 — a position he described as his boldest prediction yet. His competitors, including other AI company founders, have publicly suggested the timeline could be even shorter.

On the IPO question, Fortune also reported that Altman said he was “0% excited” to be the CEO of a public company, even as OpenAI continues taking structural steps that would be prerequisites for a public offering. The tension between Altman’s stated preferences and OpenAI’s organizational trajectory — the company converted to a for-profit structure in 2025 — remains unresolved publicly.

What This Means

The Samsung deal is the clearest signal yet that OpenAI’s enterprise business is scaling into the tier of contracts that can materially affect revenue. A deployment covering an entire division of one of the world’s largest electronics manufacturers, across R&D, manufacturing, and marketing, is a different category of commitment than pilot programs or departmental rollouts.

The Daybreak expansion and GPT-5.5-Cyber release show OpenAI moving deliberately into cybersecurity as a vertical — not just as a general capability, but with purpose-built tooling and benchmark performance to back it. The 85.6% CyberGym score gives security teams a concrete number to evaluate against existing tooling.

Altman’s “0% excited” comment about the IPO is worth taking at face value as a signal about internal culture, even if it doesn’t change the structural reality. OpenAI’s conversion to a for-profit entity and its fundraising history make a public offering a logical endpoint; Altman’s reluctance may shape timing and structure more than it shapes outcome.

FAQ

What is ChatGPT Enterprise and how does it differ from standard ChatGPT?

ChatGPT Enterprise is a version of ChatGPT designed for organizational deployments, with added data protection, user and access management, and security controls. According to OpenAI, it allows companies like Samsung to use AI within their existing governance and security frameworks, unlike the consumer product.

What is GPT-5.5-Cyber and who can access it?

GPT-5.5-Cyber is OpenAI’s security-focused model, released in full as part of the Daybreak initiative. Access is limited to verified defenders, and the model scored 85.6% on the CyberGym benchmark, which tests an agent’s ability to reproduce known vulnerabilities, per SecurityWeek.

When does Sam Altman predict AI will surpass human intelligence?

Altman has predicted AI will surpass human intelligence by 2030, according to Fortune. He described this as his boldest public prediction on the topic, and some rival AI company founders have suggested the timeline could arrive even sooner.