ECONOMY & WORK
MONEY 101
NEWS
PERSONAL FINANCE
NET WORTH
About Us Contact Us Privacy Policy Terms of Use DMCA Opt-out of personalized ads
© Copyright 2023 Market Realist. Market Realist is a registered trademark. All Rights Reserved. People may receive compensation for some links to products and services on this website. Offers may be subject to change without notice.
MARKETREALIST.COM / NEWS

Introducing OpenAI Sora, a Tool That Will Transform Text Prompts Into Real-Life Videos

Sora can also compile multiple shots in a single video giving creators to experiment in their own space.
PUBLISHED FEB 16, 2024
Cover Image Source: The Open AI logo is displayed on mobile | Pixabay | Photo by Mojahid Mottakin
Cover Image Source: The Open AI logo is displayed on mobile | Pixabay | Photo by Mojahid Mottakin

We have heard everyone say AI is the future and Sam Altman is working relentlessly to make this a true story. Open AI caught the spotlight with the Chat GPT release and is all set to utilize AI to take it to newer heights. The company has recently released OpenAI Sora, a tool that will help users create high-quality videos based on text descriptions. The recent announcement has stunned the users as everyone is still deciphering what more AI has to offer. Everyone is just going crazy to test and see what the new AI tool has in store. The company further stated that the tool is right now undergoing red teaming. This security exercise is used by companies to test the product as a competitor and point out the red flags or anything that can be improved. Let's delve deeper into the working mechanism of the tool.


 
 
 
 
 
View this post on Instagram
 
 
 
 
 
 
 
 
 
 
 

A post shared by COMPLEX (@complex)


 

The company aims to use generative AI to teach human reflexes and the elements of the physical world. Since the release of ChatGPT, Open AI has integrated GPT-4, and DALL-E 3 versions making human life and tasks easier. With Sora, they are focusing on making deep interactions with humans by making high-resolution videos just based on the text prompts. The videos can be up to one minute long with no disruption in the quality and requirements. The AI tool can not just create videos but also give users scenarios with multiple characters ensuring to understanding of the existence of elements in physical dynamics.



 

The model is trained in such a way that it completely understands the language prompts and gives the users multiple results based on the prompts. Additionally, Sora can also compile multiple mind-blowing shots in a single video giving creators to experiment in their own space. Artists such as filmmakers, designers, photographers, and other visual artists have been provided access to the tool to get feedback and point out the needed improvement areas. The app is currently facing language and prompt challenges as it does not fully understand the details and the results might not be accurate. The model is seen battling with left and right movements in some instances. 

Every app owner is concerned about the safety of the product they create. Hence, OpenAI is working continuously with red teamers to ensure safe usage of the Sora tool. The company says that it is working with industry experts to avoid the creation of hateful and misrepresented content. It is crucial to test the model in all its phases, considering the worst-case scenarios too. Further OpenAI plans to integrate a detection classifier that will easily tell the user whether or not the video is created by Sora. The company plans to amalgamate C2PA metadata just in case they release the tool as an OpenAI product.



 

Sora is not the same as DALL-E 3 but the safety measures used for the latter are considered for the new tool. They are working with the policymakers to ensure that their usage policies are safe and secure. Furthermore, they are integrating text classifiers which will analyze the text prompt and reject if it contravenes with the usage policy. Such requests might include prompts related to disturbing images, sexual content, and other violence-related imaginations. They are also working to include robust image classifiers that will first review the image frames and verify whether those go in line with the policy pointers. The work on incorporating newer safety measures is still going on as the app's safety is the top concern for the company.

MORE ON MARKET REALIST
While Steve Harvey was eager to start the Fast Money round, the players were busy dancing.
9 hours ago
The woman shared that one package had a single banana and an industrial size box of ramen.
10 hours ago
Seasoned viewers took to the internet to call out the blunder while the woman seemed unaware.
11 hours ago
His mother also joined him in celebration and walked right to the car.
12 hours ago
Belz later said that nothing is as good as the win and it was his dream to be on the show.
1 day ago
Her technique was unusual but it did the trick and everyone was elated about the spectacular win.
1 day ago
Seeing others win big may be exciting but not many people quite the tax burden that follows.
2 days ago
The host was asked about the possibility of four $100,000 wins in a row, but said that it was virtually impossible.
2 days ago
The shopper was alarmed after seeing what she thought of as objects crawling around.
2 days ago
The player had proceeded to the bonus round after simply adding one letter to another contestant's guess.
3 days ago
Steve Harvey's facial expressions are gold especially when he is geuinely shocked by a contestant.
3 days ago
The host regretted celebrating the answer as it could put his relationship in jeopardy.
3 days ago
Plinko is believed to be a game of luck by many but Carey might know a few things that we don't.
3 days ago
She was again approached by the employee who had first stopped her, and hoped that the one who helped her won't be fired.
4 days ago
Turns out the player earned the quick-witted host's respect with a shared love for football.
4 days ago
Turns out the host of Family Feud is a stickler for the rules and went on to roast the woman.
4 days ago
Viewers and the host were astonished to see just how Veonn finished with the exact sum she needed.
5 days ago
Mark Cuban said he could be the bigger person and let it go, but added that he wasn't.
5 days ago
Zoe Grobman who appeared on the show's Second Chance Tournament, addressed the issue on Reddit.
5 days ago