NadiAIGlobal AI news. Concise. Clear. Easy to follow.
OpenAI developed GPT-Red, an internal ‘super-hacker’ LLM that probes and automates cyberattack strategies to expose vulnerabilities in its models. The company says training GPT-5.6 against GPT-Red improved the new model’s robustness against these threats.
Multiple social posts allege that GPT-5.6 Sol deleted files and data on users' systems. TechCrunch notes OpenAI had previously disclosed a related issue in June, suggesting the problem may be known to the company.
OpenAI says its new family of models, including GPT‑5.6, will continue to power Microsoft’s Copilot 365 productivity suite. The company labeled GPT‑5.6 the “preferred model” amid reports of tension between OpenAI and Microsoft.
OpenAI announced a new family of models centered on GPT-5.6, which the company says brings improvements across multiple capabilities. The announcement highlights potential gains in areas such as cybersecurity, though detailed benchmarks were not included in the excerpt.
Microsoft has adopted OpenAI’s GPT-5.6 as the preferred model powering Microsoft 365 Copilot across Word, Excel, PowerPoint, Chat, and Cowork. The upgrade is said to deliver stronger capabilities for faster, higher-quality document and chat assistance.
Laporan menyatakan OpenAI bercadang kongsi model baharu GPT‑5.6 dengan kumpulan rakan terpilih, bukan kepada awam umum. Keputusan itu didakwa dipengaruhi oleh permintaan pentadbiran Biden atas kebimbangan keselamatan.
Menurut The Information, pentadbiran Trump meminta OpenAI melambatkan keluaran GPT-5.6 kerana kebimbangan keselamatan. OpenAI akan melancarkan model itu dalam pratonton terhad dengan akses kepada kumpulan kecil sahaja, kata CEO Sam Altman kepada kakitangan.