OpenAI Sora’s Rival video-focused generative AI systems continue to make waves. This time Runway Gen-3 Alpha made his debut.
Introduced last month, the Runway Gen-3 Alpha A service that can create videos from text, like OpenAI’s Sora. The system can produce high-quality videos that are 10 seconds long, particularly focusing on high realism.As of today via the link here With Gen-3 Alpha opened to everyone Here or Here impressive content that you can see can be created. Before that, Luma AI’s new system Dream Machine The system had arrived. You can watch the promotional video below, It can now produce high quality results in 5 seconds within 2 minutes.. The technology, which is stated to be able to produce longer content over time, greatly impresses those who try it. Some examples produced with the system Here And Here Dream Machine is different from many of its counterparts. here You can currently try it for free on the website.
YOU MAY BE INTERESTED IN
Last month, Kuaishou’s KLING was also announced. Able to create videos from written content and as you can see in the X share here KLING, which has shown very good results, is developed by China-based Kuaishou. The service is said to be able to create videos of up to 2 minutes at 1080P resolution and 30 fps.has the potential to create content that is definitely not AI-generated in the future. Official website Here The service is not yet available to everyone.
OpenAI’s Sora has been making headlines with YouTube details in recent months. OpenAI CTO Mira Murati asked by Joanna Stern last month, “Was Sora trained with YouTube videos?” After thinking about the question for a while, surprisingly, “I’m not entirely sure about that.” your answer had given.
Murati stated that they used publicly available videos and also specially licensed video sets in the training of the system. In order to avoid any legal problems, they do not accept it directly.interestingly enough, does not deny it, but it is very, very likely that Sora’s training was aided by millions of videos on YouTube.
It seems unlikely that the system would have reached its breathtaking current state without the support of videos on YouTube.OpenAI generally uses publicly available content as a basis for LLM (The Big Language Model) thinks that it can be used for free in education. This issue came up again later because Brad Lightcap, COO of OpenAI also directed towards, “Was Sora trained with YouTube videos?” He avoided answering the question.
Speaking on this issue, YouTube CEO Neal Mohan saidIt’s not known whether OpenAI used YouTube videos to train Sora, but stating that such a situation would be a clear violation of the platform’s policies. YouTube will file a serious lawsuit against OpenAI regarding this issue in the coming period. can open. If you missed it, the company’s new artificial intelligence model “Sora”, can produce 60-second videos from what is written And It produces very, very good results.