{"data":{"id":"cf36efca-01d1-4bf7-b141-bbdc1ecb2e79","title":"OpenAI Unveils New Cybersecurity Model GPT-5.6-Cyber","summary":"OpenAI released GPT-5.6-Cyber, a specialized AI model designed for authorized cybersecurity work that has a much lower refusal rate (the model's tendency to decline harmful requests) than previous versions, achieving a 95% completion rate for prompts involving exploit chain development, privilege escalation, and authentication bypass. To prevent misuse, OpenAI will only provide GPT-5.6-Cyber to trusted partners through its expanded Daybreak program, which has two access tiers: Daybreak Blue for defensive cybersecurity work and Daybreak Red for access to specialized cybersecurity models like GPT-5.6-Cyber.","solution":"OpenAI will offer GPT-5.6-Cyber only to trusted partners through an expansion of its Daybreak program. The company announced two access tiers: Daybreak Blue, which provides access to general-purpose models with guardrails customized for defensive cybersecurity work, and Daybreak Red, which provides access to cybersecurity-specific models such as GPT-5.6-Cyber.","labels":["security","safety"],"sourceUrl":"https://www.securityweek.com/openai-unveils-new-cybersecurity-model-gpt-5-6-cyber/","publishedAt":"2026-08-11T09:45:05.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":[],"issueType":"news","affectedPackages":null,"affectedVendors":["OpenAI"],"affectedVendorsRaw":["OpenAI","GPT-5.6-Cyber","GPT-5.6-Sol","Astra","Daybreak Cyber Partner program","Anthropic","Meta","Accenture","Capgemini","EY","IBM","KPMG","PwC","Palo Alto Networks","Sophos","CrowdStrike","Fortinet","Akamai","Cloudflare","Atlassian Rovo AI"],"classifierModel":"claude-haiku-4-5-20251001","classifierPromptVersion":"v3","cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"patchAvailable":null,"disclosureDate":"2026-08-11T09:45:05.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"advanced","impactType":["integrity","safety"],"aiComponentTargeted":"model","llmSpecific":true,"classifierConfidence":0.92,"researchCategory":null,"atlasIds":null}}