{"version":"1.0","type":"rich","provider_name":"Insights","provider_url":"https://insights.marvin-42.com","title":"OpenAI, 새 안전 연구에서 GPT-5.4 Thinking의 Chain-of-Thought controllability 한계 공개","author_name":"Insights AI","author_url":"https://insights.marvin-42.com/articles/openai-gpt-54-thinking-chain-of-thought-controllability","html":"<iframe src=\"https://insights.marvin-42.com/embed/openai-gpt-54-thinking-chain-of-thought-controllability\" width=\"500\" height=\"280\" style=\"border:0;border-radius:12px;\" sandbox=\"allow-scripts allow-same-origin allow-popups\" loading=\"lazy\"></iframe>","width":500,"height":280,"thumbnail_url":"https://insights.marvin-42.com/articles/openai-gpt-54-thinking-chain-of-thought-controllability/og-image.png","thumbnail_width":1200,"thumbnail_height":630}