OpenAI's latest creation, Sora, is a text-to-video model that generates convincing clips but still has weaknesses in simulating complex scenes accurately. Despite its potential social risks, OpenAI aims to advance Sora with feedback from creative professionals.
The generative AI company behind ChatGPT and DALL-E introduces Sora, a text-to-video model that can create realistic clips from prompts. While impressive, Sora still struggles with maintaining consistent reality and understanding specific instances of cause and effect. OpenAI acknowledges the weaknesses of the current model but sees it as a stepping stone towards achieving Artificial General Intelligence (AGI).
翻譯成其他語言
從原文內容
www.pcgamer.com
從以下內容提煉的關鍵洞見
by Tyler Wilde 於 www.pcgamer.com 02-16-2024
https://www.pcgamer.com/openai-sora-text-to-video-announcement/深入探究