OpenAI has launched Sora 2, a new tool for making videos and audio. The company says this version is better than the first Sora because it makes videos look more real, more accurate, and easier to control. On Tuesday, OpenAI also released a new social app called Sora, where people can create short videos of themselves and friends and share them on a feed similar to TikTok.
Earlier, Wired had reported that OpenAI was working on a social platform like this, and now it is here.
Sora 2 Vs Veo 3: Which One Is The Best?
An Instagram page calledGet Into AI posted clips comparing Sora 2 with Google’s Veo 3. The videos showed big differences. For example, in one video, a man was riding a horse. The Sora 2 clip looked smooth and natural, but the Veo 3 version looked over the top, showing the man bouncing strangely on the horse.
We don’t know if the same prompts were used or if they were a little different. But if we trust what was posted, Sora 2 gave the better result this time.
How OpenAI Is Different From Sora 2
Sora 2 is made to build videos that look very real and follow natural rules like physics. It also helps people make longer clips with many shots and even lets them add their own face and voice to the video.
Google’s Veo 3 works differently. It is made to connect directly with YouTube and the Gemini API. This makes it easy and fast for creators to make Shorts and upload them right away.
While Sora 2 is about longer, story-like videos with sound that matches the scene, Veo 3 is about speed, quick editing, and good quality.
Both tools care about safety. Sora 2 uses checks to stop misuse, while Veo 3 adds watermarks and uses YouTube tools to block fake content.
