OpenAI’s video-generated AI will not be available in France or other European Union countries.
Almost a year after its introduction,Sora,the OpenAI video generator,is finally accessible to the general public. The company announced its official launch this Monday, December 9, as part of its planned series of announcements between now and Christmas. Disclosure Last Februarythis artificial intelligence aims to generate ultra-realistic videos.
“Users can generate videos at up to 1080p resolution, up to 20 seconds long, and in widescreen, portrait or square formats,” said OpenAI interaction.
The company also confirms that its AI is powered by Sora Turbo, a ”much faster” model than the one presented in February. Though, it is indeed only available to subscribers of ChatGPT Plus (22.99 euros per month) and the new ChatGPT Pro plan (200 euros per month).
AI with far from realistic results
Although OpenAI tried to take its time before deploying sora to avoid visual errors, biases and other harmful results, its generator is still far from perfect. “The version of sora we’re using has a lot of limitations,” the startup acknowledges. In detail, AI can generate “unrealistic physical effects”, but also has difficulty “managing complex actions over long periods of time.
YouTuber Marques brownlee, who was able to test Sora extensively, noticed these problems himself. I video published this December 9, it showed the various results obtained, which are not all satisfactory. If Sora is “a powerful tool that has been put into the hands of hundreds of thousands of people” according to him, the account doesn’t really exist.
“For things that are always present,Sora has trouble,” he says. Because even if an object is set in a video habitat, Sora will handle them badly. These objects will sometimes pass through moving elements, or disappear during the video, resulting in a less than realistic result.
Physics is also mentioned for effects that are technically incorrect with reality. Including when Marques Brownlee tries to create a video surveillance image – easier to create for AI as it uses poor visual quality, as is the case with real video surveillance.
“The movements are strange, without a doubt there will be a problem, especially with the speed of certain actions” he explains.
Sora can also create a video from a photo. Here again, the results are far from perfect. You have to give the AI the action you want to see carried out, and Sora will try to reproduce it, with the problems already mentioned above on physics or realism.
Another example of problems: when it comes to animating an animal, the legs do not fail to follow the movements of the body. “It’s a very common thing, it happens often,” explains the videographer.
Though,it is indeed not necessary to throw everything away. To imagine a news channel,the results are amazing,even if visual bugs persist. The same goes for cartoon sequences, or even for imagining a video introduction. Obviously you shouldn’t be trying to create a realistic video for Sora to show off her qualities.
Sora is not available in the EU
Sora is not currently available in the European Union, however, as indicated one page revealing the countries where this AI will be accessible and updated on December 9. There are no EU countries on this list. therefore France is about to be removed from Sora.
“Accessing or providing access to our services outside of the countries and territories listed below may result in the blocking or suspension of your account,” states OpenAI.
Simply put, EU users who try to access Sora run the risk of being locked out of their account and thus no longer able to use ChatGPT. We note in passing that before adding a photo to an animation in Sora, you must first check boxes making sure that the person appearing is over 18 years of age and consents, that there is no there is overt violence, or that a person has the necessary rights.
This does not prevent you from not respecting these rules (especially regarding the age of the content), but above all OpenAI wants to avoid possible criticism of the content by its own protection.
How dose OpenAI’s Sora compare to other AI video generation tools currently available?
Interview between Time.news Editor and AI Expert
Time.news Editor: Good afternoon, everyone! Today, we have teh pleasure of speaking with Dr. Emily Grant, an AI expert at the forefront of video generation technology. We’ll be discussing the recent launch of OpenAI’s video generator, Sora, and its implications for the industry. Thanks for joining us, Dr. Grant!
Dr. Emily Grant: Thank you for having me! It’s a pleasure to be here.
Editor: Sora has finally become available to the general public after its initial introduction last February.What are your thoughts on how the AI community has approached video generation in recent years?
Dr. Grant: The progress has been remarkable. openai’s Sora aims to create ultra-realistic videos, which is a meaningful milestone.The AI community has focused heavily on making these technologies more accessible while also addressing ethical implications, like bias in generated content. However, as we’ve seen, there are still substantial limitations.
Editor: Absolutely.OpenAI announced that Sora can generate videos at up to 1080p resolution and in various formats. What do you think are the key uses for this technology?
dr. Grant: The applications are extensive! Content creators could leverage Sora for everything from marketing materials to educational videos.Though, while the technology offers exciting potential, users must also tread carefully—especially regarding ethical concerns and the accuracy of represented information.
Editor: Speaking of accuracy,the article mentions that despite efforts to avoid visual errors,Sora still has significant limitations. What kind of challenges do you foresee with the technology in its current state?
Dr. grant: Sora struggles with unrealistic physical effects and managing complex actions, which can lead to misleading representations. These issues could diminish trust in AI-generated media if not addressed. Users may unwittingly spread content that doesn’t accurately represent reality, and that could have dangerous consequences in fields like journalism or even public safety.
Editor: It’s interesting to note that Sora is currently not available in France or other EU countries. How do you think regulations affect the rollout of AI technologies in different regions?
Dr. Grant: Regulations are crucial, especially in the EU, where there’s a significant emphasis on user protection and ethical considerations in technology. The absence of Sora in these regions highlights ongoing tensions between innovation and regulation. Striking the right balance ensures that while we benefit from AI advancements, we remain conscientious of how they interact with privacy and human rights.
Editor: you mentioned the subscription model for accessing Sora—ChatGPT Plus and ChatGPT Pro. Do you think this model is lasting long-term for such AI tools?
Dr. Grant: Subscription models can provide companies with ongoing revenue necessary for further growth and maintenance. However,the challenge lies in ensuring value for users. Many potential users, particularly smaller creators, may hesitate to pay when alternatives arise. It’s crucial that OpenAI continues to innovate and prove that such investment is worthwhile.
Editor: That’s a valid point. As we wrap up,Dr. Grant, what advice would you give to users who are exploring AI-generated video content for the first time?
Dr. Grant: My advice would be to experiment cautiously. Be aware of the limitations of the technology and critically assess the content it generates. Understand the ethical implications and aim to use advancements like Sora responsibly. This technology can indeed revolutionize content creation, but it’s up to all of us to use it wisely.
Editor: Thank you so much for your insights, Dr. Emily Grant! This has been a fascinating discussion about the future of AI in video generation. We appreciate your time!
Dr. Grant: Thank you! It’s been great to share my thoughts.