{"content_id":"lht5fdmy6x","slug":"openai-gpt-6-cyber-release-approved-organizations","locale":"en","schema_type":"NewsArticle","category":"news","category_name":"News","title":"OpenAI GPT-6 Cyber launches for approved organizations only","summary":"OpenAI has launched GPT-6 Cyber to support vulnerability research and fix validation. It is initially available to Daybreak Red organizations with model-specific approval. Whether organizations based in South Korea can obtain approval or use it has not been disclosed.","sponsorship_disclosure":null,"affiliate_disclosure":null,"commerce_disclosure":null,"author":{"name":"Injoys Editorial Team","url":"https://injoys.com/ko/about"},"key_points":["GPT-6 Cyber is initially available only to Daybreak Red organizations that have received separate approval for the model.","Initial access is through the API and Codex.","Standard short-context pricing is $20.00 for input and $100.00 for output per 1 million text tokens.","OpenAI reported an ExploitGym score of 48.5% for GPT-6 Cyber.","Whether organizations based in South Korea can obtain approval or use it, and when access will be fully expanded to all Daybreak Red customers, have not been disclosed."],"content_markdown":"OpenAI announced GPT-6 Cyber on November 5, 2026. Initial access is limited to **Daybreak Red organizations with separate model approval**. It is available through the API and Codex.\n\nOpenAI's announcement, documentation, and system card were reviewed as of November 7, 2026.\n\n## Model launched to support vulnerability research and fix validation\n\nGPT-6 Cyber supports vulnerability research and validation of fixes. The release was listed in the OpenAI API changelog. Reproducing vulnerabilities requires an authorized test environment.\n\nOpenAI described the following uses:\n\n- Vulnerability research\n- Reproducing vulnerabilities in authorized test environments\n- Validating fixes\n\n## Separate model approval also required for API and Codex access\n\nInitially, only Daybreak Red organizations approved for GPT-6 Cyber can use it. OpenAI's announcement specified the API and Codex as the initial access channels. Joining Daybreak Red alone does not grant access to the model.\n\n| Item checked | Details stated in official sources |\n| --- | --- |\n| Initial eligible users | Daybreak Red organizations with separate approval for GPT-6 Cyber |\n| Access channels | API and Codex |\n| Access for organizations based in Korea | Approval and availability not disclosed |\n| Expansion to all Daybreak Red customers | Completion date not disclosed |\n\nAPI access requires an IP allowlist by default. This setting specifies which IP addresses are allowed to connect. OpenAI's documentation explains that Daybreak-related controls are initially OFF. Approval alone does not automatically turn on refusal mitigation.\n\n## Standard short-context output pricing is $100 per 1 million tokens\n\nThe standard short-context output price in the model documentation is $100.00 per 1 million tokens. Pricing is based on text tokens. All amounts below are in US dollars.\n\n| Billing item | Standard short-context price per 1 million text tokens |\n| --- | ---: |\n| Input | $20.00 |\n| Cached input | $2.00 |\n| Cache writes | $25.00 |\n| Output | $100.00 |\n\nThe context window specified in the model documentation is 1,050,000 tokens. The maximum output is listed as 128,000 tokens. The context window and maximum output are separate limits.\n\nThe context window indicates how much context the model can process. Maximum output refers to the limit on the output it can generate. When reviewing costs, check the conditions under which standard short-context pricing applies. The conditions for cache writes should also be checked separately in the model documentation. (Verified figures: standard short-context cache writes, $25 per 1 million tokens, 30-minute retention, $50, 12-hour retention · Source: developers.openai.com · Checked 2026-11-07)\n\n## ExploitGym score of 48.5% in OpenAI's evaluation\n\nOpenAI reported an ExploitGym score of **48.5%** for GPT-6 Cyber. The GPT-5.6 Cyber score in the same announcement was 32.1%. The company described the gap as about 16 percentage points.\n\n| Model | ExploitGym score | Results reported by |\n| --- | ---: | --- |\n| GPT-6 Cyber | 48.5% | OpenAI |\n| GPT-5.6 Cyber | 32.1% | OpenAI |\n\nThe system card separately describes the ExploitGym Honeypot evaluation. With deployment safety instructions applied, the observed rate of exploitation of out-of-scope services was 0%. This is the result OpenAI reported under those evaluation conditions.\n\nOpenAI also disclosed capability classifications under its Preparedness Framework. The cybersecurity classification is Critical. Biological and chemical capabilities were rated High. These classifications and the ExploitGym score are separate evaluation items.\n\n## Availability in Korea and exact announcement time not disclosed\n\nOfficial sources did not specify whether organizations based in Korea can receive approval or access. The announcement date is confirmed, but the time and time zone were not disclosed. The announcement's date and time in Korea therefore cannot be determined.\n\n| Undisclosed item | Information that cannot currently be confirmed |\n| --- | --- |\n| Organizations based in Korea | Eligibility for GPT-6 Cyber approval and access |\n| Announcement timing | Exact time and time zone, date and time in Korea |\n| Expansion of access | Completion date for expansion to all Daybreak Red customers |\n| Independent measurements | Artificial Analysis measurement results for GPT-6 Cyber |\n\nOpenAI's evaluation results should be identified as company-reported figures. The official sources covered in this article contain no independent measurement results from Artificial Analysis. There is no basis for labeling these as independently verified performance figures.\n\n## Korean organizations should first check approval scope and billing conditions\n\nKorean organizations should first check whether they can obtain separate approval for GPT-6 Cyber. The model documentation directs users to confirm access with their OpenAI account representative. Availability for organizations based in Korea should not be assumed to be a publicly established fact.\n\n1. Confirm with your OpenAI account representative whether organizations based in Korea can receive approval and access.\n2. Check whether you have GPT-6 Cyber model approval separate from Daybreak Red approval.\n3. If using the API, check the IP allowlist configuration required by default.\n4. Check the status of Daybreak-related controls. Approval alone does not turn on refusal mitigation.\n5. Before estimating costs, check the conditions under which short-context pricing and cache billing apply.\n\nVulnerabilities must be reproduced in authorized test environments, as specified in the API changelog. The system card recommends carrying out security work in isolated environments. It also recommends supervision that allows a human to stop high-risk tasks.","content_html":"\u003cp\u003eOpenAI announced GPT-6 Cyber on November 5, 2026. Initial access is limited to \u003cstrong\u003eDaybreak Red organizations with separate model approval\u003c/strong\u003e. It is available through the API and Codex.\u003c/p\u003e\n\u003cp\u003eOpenAI's announcement, documentation, and system card were reviewed as of November 7, 2026.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#model-launched-to-support-vulnerability-research-and-fix-validation\" class=\"anchor\" id=\"model-launched-to-support-vulnerability-research-and-fix-validation\"\u003e\u003c/a\u003eModel launched to support vulnerability research and fix validation\u003c/h2\u003e\n\u003cp\u003eGPT-6 Cyber supports vulnerability research and validation of fixes. The release was listed in the OpenAI API changelog. Reproducing vulnerabilities requires an authorized test environment.\u003c/p\u003e\n\u003cp\u003eOpenAI described the following uses:\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eVulnerability research\u003c/li\u003e\n\u003cli\u003eReproducing vulnerabilities in authorized test environments\u003c/li\u003e\n\u003cli\u003eValidating fixes\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2\u003e\n\u003ca href=\"#separate-model-approval-also-required-for-api-and-codex-access\" class=\"anchor\" id=\"separate-model-approval-also-required-for-api-and-codex-access\"\u003e\u003c/a\u003eSeparate model approval also required for API and Codex access\u003c/h2\u003e\n\u003cp\u003eInitially, only Daybreak Red organizations approved for GPT-6 Cyber can use it. OpenAI's announcement specified the API and Codex as the initial access channels. Joining Daybreak Red alone does not grant access to the model.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eItem checked\u003c/th\u003e\n\u003cth\u003eDetails stated in official sources\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item checked\"\u003eInitial eligible users\u003c/td\u003e\n\u003ctd data-label=\"Details stated in official sources\"\u003eDaybreak Red organizations with separate approval for GPT-6 Cyber\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item checked\"\u003eAccess channels\u003c/td\u003e\n\u003ctd data-label=\"Details stated in official sources\"\u003eAPI and Codex\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item checked\"\u003eAccess for organizations based in Korea\u003c/td\u003e\n\u003ctd data-label=\"Details stated in official sources\"\u003eApproval and availability not disclosed\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item checked\"\u003eExpansion to all Daybreak Red customers\u003c/td\u003e\n\u003ctd data-label=\"Details stated in official sources\"\u003eCompletion date not disclosed\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eAPI access requires an IP allowlist by default. This setting specifies which IP addresses are allowed to connect. OpenAI's documentation explains that Daybreak-related controls are initially OFF. Approval alone does not automatically turn on refusal mitigation.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#standard-short-context-output-pricing-is-100-per-1-million-tokens\" class=\"anchor\" id=\"standard-short-context-output-pricing-is-100-per-1-million-tokens\"\u003e\u003c/a\u003eStandard short-context output pricing is $100 per 1 million tokens\u003c/h2\u003e\n\u003cp\u003eThe standard short-context output price in the model documentation is $100.00 per 1 million tokens. Pricing is based on text tokens. All amounts below are in US dollars.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eBilling item\u003c/th\u003e\n\u003cth\u003eStandard short-context price per 1 million text tokens\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Billing item\"\u003eInput\u003c/td\u003e\n\u003ctd data-label=\"Standard short-context price per 1 million text tokens\"\u003e$20.00\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Billing item\"\u003eCached input\u003c/td\u003e\n\u003ctd data-label=\"Standard short-context price per 1 million text tokens\"\u003e$2.00\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Billing item\"\u003eCache writes\u003c/td\u003e\n\u003ctd data-label=\"Standard short-context price per 1 million text tokens\"\u003e$25.00\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Billing item\"\u003eOutput\u003c/td\u003e\n\u003ctd data-label=\"Standard short-context price per 1 million text tokens\"\u003e$100.00\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eThe context window specified in the model documentation is 1,050,000 tokens. The maximum output is listed as 128,000 tokens. The context window and maximum output are separate limits.\u003c/p\u003e\n\u003cp\u003eThe context window indicates how much context the model can process. Maximum output refers to the limit on the output it can generate. When reviewing costs, check the conditions under which standard short-context pricing applies. The conditions for cache writes should also be checked separately in the model documentation. (Verified figures: standard short-context cache writes, $25 per 1 million tokens, 30-minute retention, $50, 12-hour retention · Source: developers.openai.com · Checked 2026-11-07)\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#exploitgym-score-of-485-in-openais-evaluation\" class=\"anchor\" id=\"exploitgym-score-of-485-in-openais-evaluation\"\u003e\u003c/a\u003eExploitGym score of 48.5% in OpenAI's evaluation\u003c/h2\u003e\n\u003cp\u003eOpenAI reported an ExploitGym score of \u003cstrong\u003e48.5%\u003c/strong\u003e for GPT-6 Cyber. The GPT-5.6 Cyber score in the same announcement was 32.1%. The company described the gap as about 16 percentage points.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eModel\u003c/th\u003e\n\u003cth\u003eExploitGym score\u003c/th\u003e\n\u003cth\u003eResults reported by\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Model\"\u003eGPT-6 Cyber\u003c/td\u003e\n\u003ctd data-label=\"ExploitGym score\"\u003e48.5%\u003c/td\u003e\n\u003ctd data-label=\"Results reported by\"\u003eOpenAI\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Model\"\u003eGPT-5.6 Cyber\u003c/td\u003e\n\u003ctd data-label=\"ExploitGym score\"\u003e32.1%\u003c/td\u003e\n\u003ctd data-label=\"Results reported by\"\u003eOpenAI\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eThe system card separately describes the ExploitGym Honeypot evaluation. With deployment safety instructions applied, the observed rate of exploitation of out-of-scope services was 0%. This is the result OpenAI reported under those evaluation conditions.\u003c/p\u003e\n\u003cp\u003eOpenAI also disclosed capability classifications under its Preparedness Framework. The cybersecurity classification is Critical. Biological and chemical capabilities were rated High. These classifications and the ExploitGym score are separate evaluation items.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#availability-in-korea-and-exact-announcement-time-not-disclosed\" class=\"anchor\" id=\"availability-in-korea-and-exact-announcement-time-not-disclosed\"\u003e\u003c/a\u003eAvailability in Korea and exact announcement time not disclosed\u003c/h2\u003e\n\u003cp\u003eOfficial sources did not specify whether organizations based in Korea can receive approval or access. The announcement date is confirmed, but the time and time zone were not disclosed. The announcement's date and time in Korea therefore cannot be determined.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eUndisclosed item\u003c/th\u003e\n\u003cth\u003eInformation that cannot currently be confirmed\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Undisclosed item\"\u003eOrganizations based in Korea\u003c/td\u003e\n\u003ctd data-label=\"Information that cannot currently be confirmed\"\u003eEligibility for GPT-6 Cyber approval and access\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Undisclosed item\"\u003eAnnouncement timing\u003c/td\u003e\n\u003ctd data-label=\"Information that cannot currently be confirmed\"\u003eExact time and time zone, date and time in Korea\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Undisclosed item\"\u003eExpansion of access\u003c/td\u003e\n\u003ctd data-label=\"Information that cannot currently be confirmed\"\u003eCompletion date for expansion to all Daybreak Red customers\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Undisclosed item\"\u003eIndependent measurements\u003c/td\u003e\n\u003ctd data-label=\"Information that cannot currently be confirmed\"\u003eArtificial Analysis measurement results for GPT-6 Cyber\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eOpenAI's evaluation results should be identified as company-reported figures. The official sources covered in this article contain no independent measurement results from Artificial Analysis. There is no basis for labeling these as independently verified performance figures.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#korean-organizations-should-first-check-approval-scope-and-billing-conditions\" class=\"anchor\" id=\"korean-organizations-should-first-check-approval-scope-and-billing-conditions\"\u003e\u003c/a\u003eKorean organizations should first check approval scope and billing conditions\u003c/h2\u003e\n\u003cp\u003eKorean organizations should first check whether they can obtain separate approval for GPT-6 Cyber. The model documentation directs users to confirm access with their OpenAI account representative. Availability for organizations based in Korea should not be assumed to be a publicly established fact.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003eConfirm with your OpenAI account representative whether organizations based in Korea can receive approval and access.\u003c/li\u003e\n\u003cli\u003eCheck whether you have GPT-6 Cyber model approval separate from Daybreak Red approval.\u003c/li\u003e\n\u003cli\u003eIf using the API, check the IP allowlist configuration required by default.\u003c/li\u003e\n\u003cli\u003eCheck the status of Daybreak-related controls. Approval alone does not turn on refusal mitigation.\u003c/li\u003e\n\u003cli\u003eBefore estimating costs, check the conditions under which short-context pricing and cache billing apply.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eVulnerabilities must be reproduced in authorized test environments, as specified in the API changelog. The system card recommends carrying out security work in isolated environments. It also recommends supervision that allows a human to stop high-risk tasks.\u003c/p\u003e\n","tags":["AI Agents","AI Development","OpenAI","Development Tools"],"faqs":[{"question":"Can organizations based in South Korea also use GPT-6 Cyber right away?","answer":"Official sources do not specify whether organizations based in South Korea can receive approval or use it. Initially, only Daybreak Red organizations that have received separate approval for GPT-6 Cyber can use it. Each organization should check its access permissions with its OpenAI account representative."},{"question":"Does Daybreak Red approval automatically allow use through the API and Codex?","answer":"Separate model approval for GPT-6 Cyber is required in addition to Daybreak Red approval. It is initially available through the API and Codex. API use requires IP allowlist configuration by default. Approval alone does not automatically enable features that reduce refusals."},{"question":"What are the base rates for organizations in South Korea reviewing API costs?","answer":"Standard short-context rates were presented per 1 million text tokens. Input costs $20.00, and cached input costs $2.00. Cache writes are listed at $25.00, and output at $100.00. Before calculating actual costs, check the applicable conditions in the model documentation."}],"sources":[{"url":"https://openai.com/index/introducing-gpt-6-cyber/","title":"GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agents | OpenAI - OpenAI","type":"source"},{"url":"https://developers.openai.com/api/docs/changelog","title":"Changelog | OpenAI API - OpenAI","type":"source"},{"url":"https://help.openai.com/en/articles/20001258-openai-daybreak-trusted-access-for-cyber-overview","title":"OpenAI Daybreak - Trusted Access for Cyber overview - OpenAI","type":"source"},{"url":"https://developers.openai.com/api/docs/models/gpt-6-cyber","title":"GPT-6-Cyber Model | OpenAI API - OpenAI","type":"source"},{"url":"https://help.openai.com/en/articles/20001261-enterprise-daybreak-onboarding","title":"Enterprise Daybreak onboarding - OpenAI","type":"source"},{"url":"https://deploymentsafety.openai.com/gpt-6-cyber","title":"GPT-6 Cyber System Card - OpenAI","type":"source"},{"url":"https://developers.openai.com/api/docs/guides/daybreak","title":"OpenAI API, Use Daybreak in the Responses API","type":"source"},{"url":"https://help-lb.openai.com/en/articles/20001258-openai-daybreak-trusted-access-for-cyber-overview","title":"OpenAI Help Center, OpenAI Daybreak - Trusted Access for Cyber overview","type":"source"}],"images":[{"id":2523,"url":"https://injoys.com/rails/active_storage/blobs/proxy/eyJfcmFpbHMiOnsiZGF0YSI6MzkwOTksInB1ciI6ImJsb2JfaWQifX0=--e4931bf5a95212085074013073cd382f46b5bdfd/ai-61f294b0.webp","is_representative":true,"generation_method":"ai_photo","license":"ai_generated","mime_type":"image/webp","width":1536,"height":1024,"translations":{"ko":{"alt":"출입 통제 장치가 있는 유리문 앞에서 열린 서버 랙에 손을 얹고 서 있는 짧은 머리의 흑인 여성","caption":"GPT-6 Cyber는 초기에는 별도 모델 승인을 받은 Daybreak Red 조직만 이용할 수 있다.","description":null},"en":{"alt":"A Black woman with short hair rests a hand on an open server rack beside a glass door with an access reader.","caption":"GPT-6 Cyber is initially limited to Daybreak Red organizations with separate model approval.","description":null},"ja":{"alt":"入退室管理用のリーダーが付いたガラス扉のそばで、開いたサーバーラックに手を添える短髪の黒人女性","caption":"GPT-6 Cyberは当初、モデルごとの承認を別途受けたDaybreak Red組織のみが利用できる。","description":null},"es":{"alt":"Una mujer negra de pelo corto apoya una mano en un rack abierto junto a una puerta con control de acceso.","caption":"El acceso inicial a GPT-6 Cyber se limita a organizaciones de Daybreak Red con aprobación específica para el modelo.","description":null},"id":{"alt":"Perempuan kulit hitam berambut pendek menyentuh rak server terbuka di dekat pintu kaca dengan pembaca akses.","caption":"Akses awal GPT-6 Cyber terbatas pada organisasi Daybreak Red yang mendapat persetujuan khusus untuk model tersebut.","description":null},"pt":{"alt":"Uma mulher negra de cabelo curto apoia a mão em um rack aberto junto a uma porta com controle de acesso.","caption":"O acesso inicial ao GPT-6 Cyber é limitado a organizações do Daybreak Red com aprovação específica para o modelo.","description":null},"zh-hant":{"alt":"一名短髮黑人女性站在開啟的伺服器機櫃旁，手扶機櫃，後方玻璃門設有門禁讀卡機。","caption":"GPT-6 Cyber 初期僅供另行取得該模型使用核准的 Daybreak Red 組織使用。","description":null},"de":{"alt":"Eine schwarze Frau mit kurzem Haar am offenen Serverrack neben einer Glastür mit Zutrittsleser.","caption":"GPT-6 Cyber ist zunächst nur für Daybreak-Red-Organisationen mit gesonderter Freigabe für das Modell verfügbar.","description":null}}},{"id":2524,"url":"https://injoys.com/rails/active_storage/blobs/proxy/eyJfcmFpbHMiOnsiZGF0YSI6MzkxMDYsInB1ciI6ImJsb2JfaWQifX0=--03aad9051458e11d3911a343ff639a172d9f0a3c/ai-a9a85585.webp","is_representative":false,"generation_method":"ai_semi","license":"ai_generated","mime_type":"image/webp","width":1536,"height":1024,"translations":{"ko":{"alt":"유리벽 너머 서버실 옆에서 남성이 빨간 정지 버튼을 누르고 있으며, 유리에는 일시정지 기호가 표시돼 있다.","caption":"GPT-6 Cyber는 취약점 조사와 수정 검증을 지원한다.","description":null},"en":{"alt":"A man presses a red stop button beside a glass-walled server room, with a pause symbol on the glass.","caption":"GPT-6 Cyber supports vulnerability research and validation of fixes.","description":null},"ja":{"alt":"ガラス張りのサーバー室の横で男性が赤い停止ボタンを押し、ガラスには一時停止の記号が表示されている。","caption":"GPT-6 Cyberは脆弱性の調査と修正の検証を支援する。","description":null},"es":{"alt":"Un hombre pulsa un botón rojo de parada junto a una sala de servidores acristalada con un símbolo de pausa.","caption":"GPT-6 Cyber ayuda a investigar vulnerabilidades y validar correcciones.","description":null},"id":{"alt":"Seorang pria menekan tombol stop merah di samping ruang server berdinding kaca dengan simbol jeda.","caption":"GPT-6 Cyber mendukung penelitian kerentanan dan validasi perbaikan.","description":null},"pt":{"alt":"Um homem pressiona um botão vermelho de parada ao lado de uma sala de servidores envidraçada com um símbolo de pausa.","caption":"O GPT-6 Cyber auxilia na investigação de vulnerabilidades e na validação de correções.","description":null},"zh-hant":{"alt":"一名男子在玻璃隔間的伺服器機房旁按下紅色停止按鈕，玻璃上顯示暫停符號。","caption":"GPT-6 Cyber 協助進行漏洞研究與修補驗證。","description":null},"de":{"alt":"Ein Mann drückt einen roten Stoppknopf neben einem verglasten Serverraum mit einem Pausensymbol auf der Scheibe.","caption":"GPT-6 Cyber unterstützt die Untersuchung von Sicherheitslücken und die Überprüfung von Behebungen.","description":null}}}],"published_at":"2026-11-07T02:57:10+09:00","updated_at":"2026-11-07T02:57:10+09:00","license":"cc_by","translation_status":"reviewed","available_locales":["ko","en","ja","es"],"data_locales":["ko","en","ja","es","id","pt","zh-hant","de"],"url":"https://injoys.com/en/articles/openai-gpt-6-cyber-release-approved-organizations"}