OpenAI's New Sora Physics Simulator Shocks The Entire World! [Text-to-Video] 🎥🤖

All right what's going on everybody in today's video I'm going to tell you absolutely everything you need to know about open ai's new text to video generation tool Sora let's dive right into [Music] it so what will we cover in today's video first we will be talking about Sora and giving a brief Sora overview then we'll cover that Tech technical capabilities next will be the challenges and ethical considerations and then industry impact will cover different Industries and how Sora impacts them

As well as different jobs and then we'll go over sora's limitations and then finally sora's role in advancing towards AGI and what the future looks like so let's Dive Right In so before I get into Sora I want to show a screenshot right here um that aged like milk as you can see here this is on Reddit three years ago someone said imagine in a few years when we can make photo realistic videos from just a few sentences AI is crazy

And it got a bunch of down votes and basically said few years LMAO text to video isn't going to happen in our lifetime bud and especially not photo realistic maybe our great grandkids might have it LOL and that one got upvoted and this is what people thought a few years ago fast forward to the day and things are a bit different and here is a tweet from Jimmy apples if you're not familiar with him in the whole AI space he is

An open AI leak and he made a tweet November 29th 2023 saying open AI made the decision to do video generation back in March this year will we see this next year reminder that they are still sitting on a new jukebox model next year as well Deep Mind Runway Peak paa and open source are going to force their hand and that's exactly what happened well not paa and and these other companies but it was mainly when Google announced their new Gemini

1.5 pro model which we'll talk about in another video more in depth once they announced that that would have been the biggest AI news in this month but then open AI dropped Sora and it blew it out of the water so Sora is a Japanese word for sky and as you can see here here is open AI announcing Sora on Twitter you can see it on the left hand side and then Sam ultman announcing Sora as well and he actually tweeted

Out saying hey guys give me some different ideas for me to generate with Sora and I'll generate them live so people on Twitter were tweeting at him giving him different ideas and I'll show some of those as well right here and he was just using Sora to generate them and it got some really crazy stuff so as you can see here he said here is Sora our video generation model today we are starting red teaming and offering access to limited number

Of creators all right so as you can see Sora is a very impressive model let's go over a quick overview of Sora so open AI Sora is a Cutting Edge AI model capable of creating realistic videos from text prompts it can animate still images extend existing videos and fill in missing frames with high accuracy and detail Sora is designed to understand and simulate the real world marking a significant step towards achieving artificial general intelligence AGI and I also want to highlight

A deleted tweet from someone who works at the Privacy department for open Ai and he said we very intentionally are not sharing it widely yet the hope is that a mini public demo kicks a social response into gear and AI safety memes tweeted out saying open AI employee explains that they released their Sora demo today to kick a social response into gear this is a wild sentence imagine any other industry saying we're about to unleash something so dangerous Society must urgently

Start prepping rumor Jimmy apples claims they've had this since last March if true what other crazy advances are they sitting on how far has AI Advanced without our knowledge and this is a great question because as we know they were sitting on GPT 4 for a long time before they actually released it to the public and now it's been almost a year of the release of GPT 4 and as we know these AI models scale exponentially so where are they at

Now what do they have in those labs at open Ai and another point is as we saw in the news if you're not if you didn't hear this Sam Alman is seeking 5 to7 trillion for a project addressing the massive Global shortage in semiconductor chips all right he wants an investment for this amount which is insane you have to think what could he have or what could he be proposing in this investment that someone would actually give him that amount of

Money could it be AGI or could it be something that's right on the cusp of AGI it makes you really think let me know in the comments down below what your thoughts are now let's cover sora's technical capabilities so Sora can generate complex scenes with multiple characters specific motions and maintain visual consistency over time capable of generating videos up to 60 seconds long featuring complex scenes camera motions and multiple characters supports a wide range of durations aspect ratios and resolutions up

To 2048 by 2048 for images diffusion trans former model so this is not the first AI video model we have open models like stable diffusion and private ones like pea P Labs Sora outperforms them all Sora is a state-of-the-art diffusion model starting with static noise like you see here on the left and refining it into detailed video and it employs a Transformer architecture for diverse visual data training it represents videos and images as collect s of smaller data units called patches

Similar to tokens in GPT models enabling it to handle a wide range of durations resolutions and aspect ratios the Transformer architecture facilitates scalability allowing Sora to generate increasingly high quality videos as training and computing power increases so you can see here at a high level we turn videos into Patches by compressing videos into lower diens dimensional latent space and then sequentially decomposing the representation into SpaceTime patches Rec captioning technique Sora leverages the rec captioning technique from di 3 generating descriptive captions

For visual training data which improves the Fidelity of generated videos to user prompts sora's physics simulation capabilities so simulating real world interactions Sora can simulate actions that affect the state of the world such as a painter leaving Strokes on a canvas or a person eating a burger and leaving bite marks Sora maintains subject consistency and temporal coherence even when subjects move out of view temporarily by giving the model foresight of many frames at a time it can also simulate digital worlds

Controlling a player in Minecraft while rendering the world and its Dynamics emergent simulation properties Sora exhibits emergent capabilities like 3D consistency long range coherence and object permanence maintaining the appearance of characters and objects even when they move out of view these properties emerge without explicit programming for 3D or object recognition suggesting that they are phenomena of scale AKA emerging properties challenges and ethic IC considerations while Sora presents numerous opportunities it also raises concerns about misinformation deep fakes and the ethical use

Of AI generated content open AI is taking safety measures by working with red teamers to test for potential harms like misinformation and bias tools are being developed to detect misleading content and Sor will incorporate safety methods from di 3 upon product integration open AI plans to engage with policy makers Educators and artists to understand concerns and identify positive use cases the company is exploring the inclusion of c2p metadata in future development to enhance content authenticity so the Coalition of content Providence

And authenticity c2p is an organization that aims to combat the spread of misinformation online by developing Tech technical standards for certifying the source and history or Providence of media content it was formed through an alliance between major tech companies such as Adobe arm Intel Microsoft and trck the c2p efforts are focused on creating an open royalty-free technical standard that can be adopted globally to ensure the authenticity of digital content now the issue with this is if someone shares this video or

Picture that is c2p encrypted within the metadata and then someone simply screenshots it or downloads it it will lose that encryption within the metadata industry impact Sora textto video technology is expected to profoundly influence content creation education marketing and entertainment it showcases the potential for AI to assist in productivity education and creative professions openingi Sora is compared favorably to other video generation models offering higher resolution and longer video generation capabilities content creation and entertainment Sora enables the creation of highquality videos

From tax revolutionizing film making TV production and digital content creation it offers a new tool for storytelling allowing creators to visualize complex scenes and narratives quickly and efficiently s's capabilities could significantly lower production costs and barriers to entry for independent filmmakers and content creators Education and Training education content can be made more engaging and accessible through s's ability to generate illustrative and dynamic videos from textual descriptions it can be used to create realistic simulations and scenarios for training purposes in various

Fields including medical safety and Technical Training marketing and advertising so marketers can leverage Sora to produce personalized and highly engaging video content fore campaigns product demos and social media content the technology Ena rapid prototyping for marketing concepts and ideas allowing for faster iteration and testing gaming and virtual reality Sora has the potential to speed up the production of game cinematics and in-game animations enhancing storytelling and immersion it could also be used to create Dynamic narrative driven content in virtual reality experiences

And simulations potential for scientific application if sar's capabilities continue to improve it could be used for scientific applications such as testing the impact of natural disasters and infrastructure or conducting chemical experiments sora's impact on different jobs the introduction of opening ey Sora a Tex to video AI tool has sparked significant discussions regarding its potential impact on various job sectors sora's capabilities to generate videos from text prompts have raised concerns about Job displacement particularly in Creative Industries while also highlighting the potential

For new job creation and the transformation of existing rules so job displacement concerns one of the primary concerns surrounding Sora is the potential for job displacement across several Industries for example entertainment industry estimates suggest a 21% reduction in US Film TV and animation jobs by 2026 due to AI technolog techologies like Sora this displacement is attributed to sora's capabilities to automate tasks traditionally performed by humans potentially leading to significant job losses in these sectors creative Fields professionals across various creative Fields

Including video editing graphic design and animation worry that Aid driven animation could render their skills obsolete all right these are people that spent either you know thousands or hundred of thousands of dollars to learn these skills or spent their whole lives trying to develop them and the fear is that the algorithms might replace human creativity and expertise causing anxiety about the future of employment in these industries so just to go over some public sentiment that we got from sores release there's

Been a lot of noise on Twitter here we have a tweet from a influencer a YouTuber moist critical a streamer he said I'm struggling to think of a single positive thing making realistic AI generated videos like this will bring it's all just net negative and dystopian and obviously he's retweeting the AI generated video of the puppies and we also have Benjamin de cranker uh he is the CEO or the founder of a company called final frame and as you can see

Here they tweeted out the creators of final frame are currently receiving anti-ai hatred messages and the site is under dos attack we are deeply disappointed to in this turn of events and doing whatever is necessary to keep the project up so he retweeted saying have been receiving hateful messages and posts since Sora went viral final frame a small project that involves two people is now being targeted hope you're happy Dells and he also tweeted out aren't stups fun and showed some

Of the messages AI will bring a dystopian nightmare you are a thick uh I'm not going to read that but you can see here you have ruined the world and he like I said he ran a project that is somewhat it's a small project that is involving Ai and video but it's nowhere obviously on the degree of Sora and he's getting these hate messages already so a lot of people are pissed off to say the least but on the other side

There will be new job opportunities and Industry transformation of course so despite the concerns Sora also presents opportunities for job creation and the transformation of existing roles AI Specialists and data analysts so Sora is at the Forefront of creating new job opportunities within the realm of AI and data analysis these rules are crucial for developing maintaining and improving AI Technologies creative empowerment through collaboration Sora can serve as a powerful tool for collaboration between humans and AI potentially leading to a higher

Quality content production this symbolic relationship could reshape R job rules with professionals evolving their skill sets to leverage AI in the creative process and then of course prompt Engineers the emergence of prompt engineering as a new career path highlights the evolving landscape of AI and creativity prompt Engineers design AI systems across various Fields indicating the potential for new job categories that blend human creativity with AI capabilities and there's going to be a lot more of course these are just a few

Now challenges and limitations within Sora so despite its capabilities Sora currently has limitations in simulating complex physics understanding cause and effect and maintaining precise spatial temporal details which are crucial for achieving AGI for example Sora might not show a bite mark on a cookie after someone takes a bite highlighting the challenges in simulating detailed physical interactions open AI acknowledges these challenges and is focused on improving sora's ability to simulate the real world accurately sword's role in advancing towards AGI so video

Generation as a path to AGI sora's ability to create realistic videos suggests that scaling video models is a promising path towards developing highly capable simulators of the physical and digital world the model's emergent capabilities in simul ating actions and digital worlds like video games indicate progress towards AGI so it's essentially learning physics all right like how chat GPT learn language this is essentially learning physics of the real world to simulate it learning physics implicitly Sora learns implicitly grasping Concepts like fluid

Dynamics and 3D Transformations which indicates an emergent understanding of physics this emergent physics understanding is demonstrated through sora's ability to simulate intricate worlds with realistic physics and here we have a tweet from Jim fan senior research scientist and Lead AI agent at Nvidia and he said if you think open AI Sora is a creative toy like dally think again Sora is a data-driven physics engine it is a simulation of many worlds real or Fantastical the simulator learns intricate rendering intuitive physics

Long Horizon reason Ing and sematic grounding all by doing den noising and gradient maths I won't be surprised if Sora is trained on lots of synthetic data using Unreal Engine 5 it has to be let's break down the following video prompt photorealistic close-up video of two pirate ships battling each other as they sail inside a cup of coffee the similat instantane Ates two Exquisite 3D assets pirate ships with different decorations so Sora has to solve text to 3D implicitly in its

Latent space the 3D objects are consistently animated as they sail and avoid each other's pass fluid dynamics of the coffee even the Foams that form around the ships fluid simulation is an entire subfield of computer Graphics which traditionally requires very complex algorithms and equations photo realism almost like rendering with Ray tracing the simulator takes into account the small size of the coffee cup compared to oceans and applies tilt shift photography to give a minuscule Vibe the semantics of the scene does

Not exist in the real world but the engine still implements correct physical rules that we expect next up add more modalities and conditioning and then we'll have a full data driven uee that will replace all the hand engineered Graphics pipelines all right all right and here we are on opening eyes official website let's check out the technical report real quick in this work we find that diffusion Transformers scale effectively as video models as well below we show a comparison of video

Samples with fixed seeds and inputs as training progresses sample quality improves markedly as training compute increases you can see here base compute and we see somewhat of a deformed dog here then we see four times the compute we see a bit more of a blurry pup with you know the owner right here um still kind of morphed and then 32 times the compute and we can see a lifelike dog a human right here painted nails boots background of trees everything on

The ground really crazy stuff and you can animate images like seen here right and extending generated videos so Sora is also capable of extending videos either forward or backward in time below our four videos that were all extended backwards in time starting from a segment of a a of a generated video as a result each of the four videos starts different from the others yet all four videos lead to the same ending we can use this method to extend a video

Both forward and backwards to produce seamless infinite Loops like you see here you know these two bikers just going down a hill in an infinite Loop and input video change the setting to be a lush jungle so you can you know change videos video and video editing connecting videos see here we have a video here and a video here and if you merge them it looks like this all right you can merge videos together which is pretty pretty cool so like

Now we merged this video and this video so you see the cheetah and the car and then you know here you see like Aqua um sharks and whatnot as well as the oldfashioned horses and everything all together in one video really cool stuff I'll leave links to all of this down in the description below you can check them out future development So currently available to a select group of testers with plans for public release after further refinement and safety assessments openingi

Aims to engage with policy makers Educators and artists to explore positive use cases and address concerns as Sora and similar Technologies evolve they could further impact Society by automating content creation tasks influencing various Industries Sora represents a significant step towards AGI by enhancing ai's ability to understand and simulate the real world and real world physics while it showcases impressive capabilities open AI is aware of its current limitations and is working to address them while considering safety and ethical implications all right

Guys so thanks for watching this video make sure to subscribe and let me know what your thoughts are in the comments down below do you think s is a good thing do you think it's a bad thing what are your thoughts on this whole topic do you think openi has AGI achieved already or do you think they're on the cusp of it with Sam asking for a $7 trillion investment we'll cover more of that stuff in future videos so make sure

To subscribe like this video also guys you'll see that bar scrolling right here at the bottom of the screen to three different websites the first is the stride Community it is a free community of like-minded AI enthusiasts entrepreneurs content creators you can go to that link stri community.com and join for free uh where there's free resources as well as you can network with like-minded individuals second is the sixday accelerate your stride AI challenge you can go to accelerate stride. comom it

Is a free challenge I will be hosting and you're going to want to you're going to want to tune in if you want to use AI in your personal life as well as your business life to really maximize things and then also too guys I just launched my new book client acquisition code and you can buy your digital copy at client acquisition code.com here is what the book looks like and it is a hefty book going over how to use AI

In your business I give specific prompts and no fluff I leave nothing out of this book digital copy now also too guys I'll leave a link to go stride level which is my custom built uh snapshot for the CRM go high level if you're not familiar with it you can check out some of the other videos on our channel it's the best CRM in the world for your business and if you sign up just simply through our affiliate link which costs

The same for you it's literally like the cheapest CRM in the market for what you get um we'll give you our custombuilt free snapshot you can learn all about it at ghostride level.com so I'll leave a link to this down below as well and you get a lot of free bonuses free training absolutely everything for free um you can just literally drag and drop it copy and paste it into your go highle account from automations workflows forms funnels you know websites

All that good stuff for free what people charge thousands and thousands of dollars for you can get simply just through signing up through our affiliate link and you'll get access to a community and a bunch of good stuff all right and you can try it out for 14 days and if you don't like it you can cancel and no harm no foul right other than that guys I will see you in the next video keep hustling keep grinding and of course

Accelerate your stride guys take care