The AI Daily Brief: Artificial Intelligence News and Analysis - Sora 2 and the Brainrot Rebellion

Episode Date: October 2, 2025

On this episode, NLW goes deep on OpenAI’s release of Sora 2—its next-generation video generation model—and the launch of the new Sora social app, which some are calling an AI-powered TikTok. Is this the beginning of a Cambrian explosion of creativity, or just the next wave of AI brain rot? The show explores what makes Sora 2 different, how the cameo feature could reshape social media, the early cultural backlash, and what this moment says about society’s growing rebellion against attention-draining digital platforms.Brought to you by:Is your enterprise ready for the future of agentic AI?⁠⁠⁠⁠⁠⁠⁠⁠Visit AGNTCY.org⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Visit Outshift Internet of Agents⁠⁠⁠⁠⁠⁠⁠⁠Try Notion AI today with Notion 3.0 ⁠⁠⁠⁠⁠⁠⁠⁠https://ntn.so/nlw⁠⁠⁠⁠⁠⁠⁠⁠KPMG – Discover how AI is transforming possibility into reality. Tune into the new KPMG 'You Can with AI' podcast and unlock insights that will inform smarter decisions inside your enterprise. Listen now and start shaping your future with every episode. ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.kpmg.us/AIpodcasts⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Blitzy.com - Go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://blitzy.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ to build enterprise software in days, not months Robots & Pencils - Cloud-native AI solutions that power results ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://robotsandpencils.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Insightwise - AI for the entire consulting lifecycle ⁠https://www.insightwise.ai/⁠Vanta - Simplify compliance - ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://vanta.com/nlw⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠The Agent Readiness Audit from Superintelligent - Go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://besuper.ai/ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠to request your company's agent readiness score.The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: https://pod.link/1680633614Interested in sponsoring the show? [e-mail]

Today on the AI Daily Brief, we are talking about the new SORA 2 model and the SORA app that goes with it, and whether it represents a potential Cambrian explosion for creativity or beginnings of the brain rot rebellion. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. Hello, friends, quick announcements before we dive in. First of all, thank you to today's sponsors, Super Intelligent, KPMG, robots and pencils, and notion. To get an ad-free version of the show, go to patreon.com. slash AI Daily Brief. And to learn about sponsorship options, send us a note at sponsors at AIDailybrief.a.i. Now, one quick additional note, this episode is a big, ponderous big thing.

We are getting into not only the new model that was released in the first reactions, but what it means for society, which means, of course, that we absolutely do not have time for headlines. We'll be back with our normal format of headlines and main episode tomorrow, but for now, let's just dive into this incredibly important conversation. This is an emergency bulletin from the national radio network. Reports from across the country tell of an alarming condition the doctors are calling AI brain rot. Across the nation, a generation has surrendered its thoughts to the artificial mind. Young and old, sit in trance.

Their brains reduced to humming circuits, a condition our doctors call AI brain rock. A new digital menace is sweeping the suburbs. It's called AI brain rock. An experts say it can hijack a teenager's imagination. Teachers report students muttering machine slang and staying up all night chatting with programs. The machine wants my head, man, it can have it. I'm just along for the ride. Why fight it?

Let it plug in, tune us up, whatever. Feels cosmic. They say it's brain rot. more like brain blossom you dig. Every single post. Another blurry dog with 16 teeth, a senator who looks like a wax figure and whatever this is supposed to be.

I didn't ask for an all-you-can-eat buffet of robot nonsense, so I'm going to need to speak to whoever's in charge of this people. We're joined this morning by the host of the AI Daily Brief, here to help us unpack the buzz around Sora and something the internet's calling BrainRot. Thanks for waking up early with us.

Glad to be here. Welcome back to the AI Daily Brief. Today's episode is about SORA. Yesterday, OpenAI made good on all the rumors and released the latest generation of their video generation model Sora. SORA 2. Interestingly, however, it wasn't just a model release. Instead, OpenAI released not only SORA 2, but also a new social app called SORA that is in some ways a TikTok or Instagram real style

short form app for sharing, collaborating on, and remixing AI videos. And if you thought the conversation around META's vibes feed from last week was contentious, it has absolutely nothing on this. In fact, I'm convinced that part of what's happening now is not just a reaction to AI, but is part of a broader reckoning of social media and the midst of an attempted societal reclamation away from digital media that we are firmly in the middle of right now. Let's talk first, though, about what was actually released. In their blog post, OpenAI sums up,

our latest video generation model is more physically accurate, realistic, and more controllable than prior systems. It also features synchronized dialogue and sound effects. There have been a couple of important moments in the history of AI-generated video. One was the original announcement of SORA, which pretty fundamentally changed people's expectations and understanding of what was possible with AI video. Although, interestingly, the total lack of access that people had to the model in the subsequent months took a lot of the sheen off of it.

In fact, in some ways, it seemed like OpenAI had just comfortably surrendered that whole side of AI generation to other people. The other people in question, of course, most notably, was Google, whose V-O-3 was in many ways the seminal chat GPT-style moment when it came to AI video. The difference with V-O-3 as opposed to every other previous model was that it integrated sound. Whereas previously you would have had to wire together some complicated workflow to generate your video in one place and your audio in another and then mash them together. Now all of that could be done in a single prompt. It turns out that that was the type of change that fundamentally unlocks massive amounts of new usage. And subsequent to that release, V-O-3 videos started showing up all over social media.

We got Harry Potter vlogworts videos, tons and tons of. of Bigfoot and Yeti vlogs, which back in June how to geek actually said prove AI's worth as a creative tool, and in general just saw for the first time, AI video really become a key part of the short form video landscape. Fast forward to this announcement, SORA 2 has those V-O3 style audio capabilities, but also what OpenAI says is hugely upgraded reality when it comes to things like physics. Indeed, if the original SORA model from back in February 24 was akin to the GPT1 moment. OpenAI writes with SORA 2,

we're jumping straight to what we think may be the GPT 3.5 moment for video. They continue, Sora 2 can do things that are exceptionally difficult and in some cases outright impossible for prior video generation models. Olympic gymnastics routines, backflips on a paddleboard that accurately model the dynamics of buoyancy and rigidity, and triple axles while a cat holds on for dear life. Continuing, they write,

prior video models are over-optimistic. They will morph objects and deform reality to successfully execute upon text prompt. For example, if a basketball player misses a shot, the ball may spontaneously teleport to the hoop. They also say that the model is a, quote, big leap forward and controllability, able to follow intricate instructions spanning multiple shots

while accurately persisting world state. And they point out it has multiple styles available to it, including realistic, cinematic, and even anime. One of the big things, though, is that SORA is explicitly trying to interact the real world in which we live and ourselves with this AI-generated video world. In short, you can put yourself, your friends, or objects from the real world into different

environments using SORA. And that in many ways is the main thesis of the new SORA app that comes alongside the model. OpenAI writes that when they first started playing with the upload yourself feature internally amongst the SORA team, they had so much fun with it that they decided to embed it in a new social app, which they are calling SORA. Not only can you drop your creations into the app, you can create and remix the videos from other people that you're following, and you can also drop yourself and your friends into videos via what they're calling cameos. Basically, you take a short video of yourself saying a set of

numbers and tilting your head in a couple different directions, and then you and anyone you've authorized can use your likeness for whatever generation they want. You have the control to say whether it's just you who can use your likeness, whether it's just people you follow, whether it's an extended network of people you follow and the people they follow, or whether it's anyone. Sam Altman's announcement of this was appropriately highfalutin. He wrote, this feels to many of us like the chat GPT for creativity moment, and it feels fun and new. There's something great about making it really easy and fast to go from idea to result and the new social dynamics that emerge.

In other words, what he's implicitly saying here is that part of the value of having the app alongside the model is the instantaneous feedback loop you get by sharing the thing, rather than just cutting it off at the moment of creation. Sam continues, creativity could be about to go through a Cambrian explosion, and along with it, the quality of art and entertainment can drastically increase. Even in the very early days of playing with Sora, it's been striking to many of us how open the playing field suddenly feels. In particular, he writes, the ability to put yourself and your friends into a video, the

team he said worked very hard on character consistency, with the cameo feature is something we have really enjoyed during testing and is to many of us a surprisingly compelling new way to connect. Now, in the first 24 hours after launch, a huge part of the feed has just been people experimenting and showing off what they can do. Rory Flynn wrote, Sora too, not what I expected. five dumb slash awesome examples. One was a recreation of Love Island,

where a woman watches CCTV-style footage of a real-life Ronald McDonald making out with a real-life Wendy, a documentary about famous memes, and a guy on a skateboard and a treadmill holding a leafblower, generated from the prompt the dumbest thing you could possibly imagine. Others were looking to more business uses. Rio Lu, the head of design at Cursor, shared a vintage style ad for the company and said,

Sora 2 one-shot at a cursor ad. Mind-blown emoji. Where are we heading into? Others were more focused on testing out the capabilities. Chase Bauer showed a video of a ball rolling down a set of planks and said absolutely baffled by how good Sora 2 is at physical consistency. I'm basically watching my SIM code execute. Others pointed out that Sora passes the glass refraction test where the reflection of an arrow drawn on the wall behind the glass

changes direction when you pour water in the glass. Some explored whether it could generate old classic video games, which it did, while others had dreams of a better ending of Game of Thrones season 8. Vresor X right. The realism in SORA 2 is on another level. The physics, the motion, the way everything interacts dynamically. It doesn't just look real. It feels real.

This is no longer text to video, it's reality to video. Absolutely mind-blowing. Peter Levels writes, you really have to give it to Open AI because SORA2 is very impressive on a lot of fronts. High-quality video model with great physics, high-quality audio in each video, high-character consistency, multiple characters in one scene, accurate character's voice, social platform attached to it.

Before today, the best AI video models were dominated by Chinese companies like ByteDance and Google with V-O3. But none of these models had great character consistency. If it was a feature at all, let alone multiple characters in one scene. Generally, you'd make a video and the face would slowly change into someone else, just not good. On top of that, Google was struggling with allowing people to upload characters, scared that it would get abused for deepfakes and just generally nerfing their model so you can't really use it for anything.

OpenAI solved that by rethinking ownership over your characters smartly with Cameo, which is essentially train yourself as an AI model, which we've all been doing in our apps for years, but in a more smart way where you can control if only you can make content with your appearance or others too. Others were having similar experience. Andrew Curran wrote, I underestimated Sora too. After using it for a while, I honestly think it's being undersold. I went in comparing it one-to-one to VO, but they are not really aiming for the same thing anymore.

This is a meme machine. It's specialized, a TikTok generator. It's also very fun. And interestingly, while previously most of the discussion we've had around these models is their implication for things like advertising and for Hollywood and the entertainment industry, which certainly there was some discussion of. For example, Javi Lopez writes, there's no way Hollywood won't be affected by this,

while showing off the creation from a prompt fight of Megamex Diesel Punk War Machines in in 1920s Europe inside of a palace banquet hall scared running people, AK DSLR Hyper Real. But mostly the discussion is about what this means for regular people, a real democratization of creativity conversation. 0.005 seconds writes, comparisons to V-O-3 are hollow. Sora-2 cameo feature is going to hit Normies like crack in the 80s.

You will have grandmas DMing videos of themselves skydiving at weddings and teens sending their mom videos of daddy hitting the gritty. It's going to win so hard. And that was certainly one side of the takes. Avarosintas writes Sauru is absurdly fun and incredible. One sentence and I've got a whole ad. Entrepreneur Andrew Wilkinson wrote, I think OpenAI just killed TikTok.

I'm already laughing my head off and hooked on my Sora feed. And now the number one barrier to posting, the ability to sing, dance, perform, edit is gone, just 100% imagination. Others aren't so sure. They see the possibility, but aren't quite clear yet if it's going to actually work. Continuing from that earlier post, Peter Levels wrote, We don't know if SOR2's social platform features will actually be used or take off, but it's a real cool experiment at trying to find a way to build a community in a more Instagram-like way.

Being able to tag your friends and then add them as multiple characters is innovative in both the social and technical aspect. Signal writes, Early guesses on Sora 2 could be wrong. Expects SORA content to flood TikTok and reels the way Vio already has. The real power might just be creator makes, downloads, and shares elsewhere. Next, creating videos of even close friends still feel strange, like Facebook tagging back in the day but way more invasive.

People are guarded about their content. Next, retention will be tricky if the feed is the main hook. I don't feel drawn to going back to the feed. Early days. Let's see if it becomes a massive surface or just another tool. early adopter Robert Scobel writes, I'm playing with Sora just like everyone else. I'm already bored, fed it about a dozen of my photos,

some are fun, but hardly blew me away. And a few refuse to run due to its lockdown nature. Can't use people, even refuse to use a photo in a museum of a suit of armor. The flower slop account writes, it won't be AI TikTok. Kids 10 to 16 don't even like AI content that much and are good at identifying slop.

The entertainment factor won't last too long, everyone will try it. It'll spawn some viral memes elsewhere, but that's about it. It won't hit TikTok-level brain rot or replicate the brain rot quality, that distinct brain rot type humor is pretty messy in a deeply human way and hard to replicate for AI. So one, TikTok, reels, shorts, stay the real threat. Two, Zora will attract AI enthusiasts, but no one's spending more than 30 minutes a day on it.

Now, if you look historically, the new social apps that succeed have almost never been just a new approach to making an old form of content. They need something that actually, in retrospect, was clearly a new form of content. When Instagram came out, yes, we had had photos before, but there had never been a feed of photos that you could filter and shape to create a particular vibe that told the story of your life. Snapchat was expiring. And it turns out there were a whole new set of use cases and interactions that required that expiring nature. TikTok, remember, wasn't just short-form videos. It was also this idea of video as meme, and it had more than a little dose of using music as a key part of it. Remember, TikTok was formed as a combination of two

different apps, the original version of TikTok plus an app called Musically, which was all about music memes. So the question will be, does AI video, where you can generate your friends and yourself in these sort of settings, actually jump to that threshold of being a fundamentally new experience, or is it just a new way to make an old type of content, i.e. short form video. Today's episode is brought to you by Super Intelligent. Now, one thing that we are having a lot of conversations with folks about is the fact that for some of you, your fiscal year is coming to end. And that means two things. One, it means planning and thinking about what you're going to do in the next year. And two, it means using up those last of budgets so you don't lose them. If you are an enterprise that

happens to find yourself in that situation, super intelligent would love to help on both fronts. We are moving increasingly towards an annual AI planning model where we map out how you can create an action map of your organization's agent opportunities that represents an executable backlog of AI and agent use cases that you can deliver on over the course the next year. Additionally, for those end-of-year budgets, we have worked out deals with a number of partners where we can pre-lock-in general implementation packages even before you figured out exactly what use cases are going to require them. If you'd like to learn more about superintelligence agent readiness audits and this new end of fiscal year plan, visit us at B-super.a.i, click get started and make sure to use the word

fiscal somewhere in the description. What if AI wasn't just a buzzword, but a business imperative. On You Can with AI, we take you inside the boardrooms and strategy sessions of the world's most forward-thinking enterprises. Hosted by me, Nathaniel Wittamore, and powered by KPMG, this seven-part series delivers real-world insights from leaders who are scaling AI with purpose, from aligning culture and leadership to building trust, data readiness, and deploying AI agents. Whether you're a C-suite executive, strategist, or innovator, this podcast is your front-row seat to the future of Enterprise AI. So go check it out at www.com. KPMG.us slash AI podcasts, or search you can with AI on Spotify, Apple Podcasts, or wherever you get your

podcasts. Today's episode is brought to you by robots and pencils. When competitive advantage lasts mere moments, speed to value wins the AI race. While big consultancies bury progress under layers of process, robots and pencils builds impact at AI speed. They partner with clients to enhance human potential through AI, modernizing apps, strengthening data pipelines and accelerating cloud transformation. With AWS certified teams across US, Canada, Europe, and Latin America, clients get local expertise and global scale. And with a laser focus on

real outcomes, their solutions help organizers work smarter and serve customers better. They're your nimble, high-service alternative to big integrators. Turn your AI vision into value fast. Stay ahead with a partner built for progress. Partner with Robots and Pencils at robots and pencils.com slash AI Daily Brief. chatbots are great, but they can only take you so far. I've recently been testing Notion's new AI agents, and they are a very different type of experience. These are agents that actually complete entire workflows for you in your style, and best of all, they work in a channel that you already know and love

because they are purpose-built Notion super users. Notion's new AI agents completely expands the range of what Notion can do. It can now build documents from your entire company's knowledge base, organize scattered information into organized reports, basically do tasks that used to take days and get them complete in minutes. These agents don't just help with work, they finish it. Getting started with building on Notion is easier than ever. Notion agents are now your very own super user to help you onboard in minutes.

Your AI teammates are ready to work. Try Notion AI for free at the link in our show notes. For others, of course, the question was not about whether the app will succeed or fail. It was whether it should exist in the first place. Some people had specific complaints. Astro Joe B went viral when he shared a source. 2 video that was CCTV footage of Sam Altman stealing GPUs at Target and added the comment, wow, it's really cool that we can use AI to make fake CCTV footage of people committing

crimes. Surely this has no downsides whatsoever. For others, the conversation is about copyright. In our previous coverage, we talked about the Wall Street Journal reporting that OpenAI was taking this very interesting approach of asking copyright holders to opt out rather than assuming their copyright by default. Smokeaway published a SORA video where a speaker plays sweater weather by the neighborhood on Spotify, in perfect reality, adding music industry is going to love this. This one is notable because I can't even actually have this playing with audio in the background or else it would be demonetized on YouTube and taken down in other places. But even if some people were worried about the harmful uses and the copyright issues,

for many, it was the very existence of the thing in the first place that was offensive. Ed Newton Rex writes, the Sora app is the worst of social media and AI. Short video app designed for addiction, literally only slop nothing else. train on other people's videos without permission. Finn Twitter account liquidity writes, Open AI is building technology that will displace millions of workers while simultaneously creating the AI slop trough humans will consume to fill the void. Eichen Jin writes, we were promised AGI, ASI, personal superintelligence.

Instead, we get infinite slot machines that turn us into dopamine-addicted zombies. Sasha Goosev reposted that and said, What? Tech companies using vague promises of world-changing innovation to lure talented young people into working on optimizing internet addiction and ad delivery? Really? Colin Frazier simply writes, I do not believe that the ability to put yourself and your friends into a video is going to bring about a Cambrian explosion for creativity. Chris Backer had maybe the simplest version of this

omnipresent tweet, Open AI in 2021, we want to cure brain cancer, open AI in 2025, we're becoming brain cancer. Even many people in the AI industry weren't thrilled about this. Demi Nicolao from Wondercraft responded to one OpenAI staffer talking about their excitement around the new SORA app and said, sorry, but how exactly are you making the world a better place? Your post is nice and eloquent, but the core message is slop is fun. We made it easy to build on each other's slop, so more slop. SORA 2 is an incredible innovation and props to you for being in the team that releases something like this to the world. Productizing it as short-form brain rot slop is basically you guys weaponizing this innovation to leverage the attention

towards growing open AI. Now, as you might imagine, the company had a sense that this was going to be part of the conversation. In their announcement post, they have a section, called launching responsibly. In it, they write, concerns about doom scrolling, addiction, isolation, and RL-SL-optimized feeds are top of mind. Here is what we're doing about it. First, they say they're giving users tools and optionality to be in control of what they see. They write using OpenAI's LLMs, we've developed a new class of recommender algorithms that can be instructed through natural language. They also say that they've built-in mechanisms to periodically pull users on their well-being, and this is already something that people

are meming about. As you're scrolling the feed every five or ten videos, it asks you a question of how it's making you feel, thumbs up or thumbs down, leading to about a thousand videos so far of Sam Altman, forcing his face into a ring video asking, how do you feel? How do you feel? They also write, by default, we show you content heavily biased towards people you follow or interact with and prioritize videos that the model thinks you're most likely to use as inspiration for your own creations. They say we are not optimizing for time spent in feed, and we explicitly designed the app to maximize creation, not consumption. Indeed, they even released an entire blog post called the SORA Feed Philosophy. The principles including optimize for creativity, put users in

control, prioritize connection, and balance safety and freedom. The app, they say, is made to be used with friends. Overwhelming feedback from testers, they write, is that cameos are what made this feel different and fun to use. You have to try it to really get it, but it is new and a unique way to communicate with people. Sam Altman also talked about this in his blog post. In the second half, he writes, we also feel some trepidation. Social media has had some good effects on the world, but it's also had some bad ones. We are all in a lot. We are all. aware of how addictive a service like this could become, and we imagine many ways it could be used for bullying. It's easy to imagine the degenerate case of AI video generation that ends up with us

all being sucked into an RL-optimized slop feed. The team has put great care and thought into trying to figure out how to make a delightful product that doesn't fall into that trap, and has come up with a number of promising ideas. We will experiment in the early days of the product with different approaches. Four of the principles he outlines are optimizing for long-term user satisfaction. Basically, when you look back over six months, users should say that their life is better for using sore than if they hadn't. The other principles he articulates, including encouraging users to control their feed, prioritizing creation, and helping users achieve their long-term goals. It's clear from Twitter slash X that a lot of OpenAI employees had to go through a whole

decision set around this. Liam writes, this was initially a tough decision. As a skeptic of short-form video and entertainment at scale more generally, I held many reservations about working on this product for fear that consumer GenAI inevitably leads to engagement-baiting attention slop, the entertainment, etc. However, after trying out the SORA app and getting to know the folks building it, I have faith, more than I thought possible, that AI video can create a truly pro-social experience, something that encourages co-creation and human connection, as opposed to vacant consumption. This is no small endeavor. It requires resisting the modus operandi and perverse incentive structure of social media today. But I believe it is possible, and I consider it a great honor to be a part of the team that

is dedicated to avoiding a present or future where AI content minds attention, nurtures addiction, and waste time. This is a first try and nowhere close to perfect. It'll take time to do this right, but the mission is worth working towards. writes John Hallman from OpenAI. AI-based feeds are scary. I won't deny that I felt some concern when I first learned we were releasing Sora 2. That said, I think the team did the absolute best job they possibly could in designing a positive experience, compared to other platforms I find myself scrolling way less and creating way more. Boaz Barak writes, I share a similar mix of worry and excitement. Sora 2 is technically amazing, but it's premature to congratulate ourselves on avoiding the pitfalls of other social media apps and

deepfakes. I'm happy with a number of aspects, visual watermarking when exporting, real humans via self-uploaded cameos with control on how your identity is used, airing on side of carefulness and content policies, emphasis on higher creation consumption ratio, and interruptions of infinite scroll, but as always, there is a limit to how much we can know before a product is used in the real world. The question, though, for so many is why even do this in the first place? What is that? What is the point. Notion founder, Exchei Kothari, who is obviously no AI skeptic given how much Notion is investing in AI, writes, why do we keep dedicating our brightest minds, billions of dollars, and the most powerful GPUs on Earth to building yet another app that optimizes for attention

decay. I was hopeful when chat GPT seemed to reclaim time from TikTok and Instagram. It felt useful, even nourishing. But now we see disposable video, same engagement treadmill, and path to ads? If even our nonprofits can't resist this gravity, what does that say about? us. Rehar Jark writes, honestly, no idea why OpenAI would spend the precious scarce compute to run an AI-generated social media. As if the LLM race wasn't CAPEX intensive enough, now enter social media. The last social media company to make it was TikTok. They spent $10 billion on marketing annually to get there. Prins thinks that it actually has to do with the path to AGI, adding the satirical quote, but why does OpenAI need to release a video model? They share a recent research paper about

V-O-3. Some of the outcomes of the paper show that V-O-3 can solve a broad variety of tasks that it wasn't explicitly trained for, which enables early forms of visual reasoning like maze and symmetry solving, with the point being that yes, maybe this is valuable on the path to AGI. For others, it's even simpler. Signal writes, unfortunately, ads fund research. Google ads lead to deep-mind, meta-ads lead to AR and VR, open-AI ads lead to possible AGI. And some point out that when it comes the valuable ad space from social media, OpenAI has some unique advantages. When I asked people to steal man why they would do this, Cody McDonald responded, Open AI doesn't have the cold start problem. The next generation is already in Chachyb-T. If they can pull them out of tired old networks

into the new social layer of the AI native era, with commerce already baked in, game over. They own the future. Stability founder Ahmad Mastock writes, in an age of infinite content, human attention is one of the few finite things. So they'll try to capture as much of it as possible. Attention is all they need. Now, I've spent a lot of time thinking about this. For long before SORA was announced, or MetaVibes was announced, I think that AI is right now inheriting a much larger set of frustrations.

It has been 30 years now since the first batch of people got dial-up internet. 25 years since the first display ads made websites available at scale for free, a little over 15 years since we got the first sets of social media. And throughout most of that time, we weren't sure how it was going to affect us. There were, yes, some very positive early indicators, I think back to the excitement around the Arab Spring and democratic movements that coordinated via social media and Web 2.0 tools, but over the last decade, there has been a growing recognition that the problems of social media were much more subtle and much more insidious.

Even among the users of these tools, the sense that they are making their lives worse is on the rise. A Pew study from earlier this year found that, 48% of teens said that social media harmed people their age. That number was up from 32% in 2022. Percentage of teens that said that social media was mostly positive had dropped from 24% down to 11%. What's interesting now is that the parents of today

are the first group of parents that en masse actually had access to and use these social media tools throughout much of their youth informative years. The parents of a decade ago didn't know what their kids were using and they didn't understand how it might be affecting them. Now, obviously, my lived experience with social media as a 41-year-old is very different from a 15- or 16-year-old who lives inside the pressure and bullying of Instagram, but I've still had an Instagram account for over a decade at this point.

In response, you're starting to see things like the Wait Until Eighth Movement, which is a pledge that groups of parents take at schools to rally their communities to all delay together in giving their kids smartphones until at least the end of eighth grade. The idea, of course, is that if all the parents, parents have the same set of rules, kids don't have to face the social pressure and ostracization of not having the technology that everyone else has because everyone in the community has made that decision together. There are tons and tons of things like this happening now, but it also feels to many like a very uphill battle. Part of the rebellion that's happening right now, this brain

rot rebellion against meta vibes, against Sora 2, is that people are sick of this medium as a medium for creative expression. If your analysis of social media is increasingly that it has been a net negative, then doing more of it, even creatively, feels like a waste. And for some, it is inescapable that even if it is super creative, it's propping up a system that's designed to simply mind people for attention. Honestly, I find myself with conflicted feelings about this. On the one hand, I would characterize myself as something of an attention libertarian. I don't really think people should get to control what other adults do or don't spend their time on. I think adults can make their own choices for themselves. But as part of that attention libertarianism, I also think people

getting loud about not wanting something or thinking an experience is bad is a form of market action. I think it's reasonable then for people to push back and say to Open AI that they don't believe that this is what they should be spending their time on. It is likewise reasonable for OpenAI to say this is going to fund all the other things that we want to do, that this is a necessary part of it, a unique opportunity for us, and we're going to do it the best we can, but we're going to do it. But people do get to vote with their feet. Ed Newton-Rex also tweeted, If you're feeling depressed about the SORA 2 release, imagine how Open AI employees who join to cure all diseases are feeling. Matt Palmer writes,

would not be surprised if we see a big wave of Open AI departures in the next month or two. If you signed up to cure cancer and you just secured post-economic bags in a secondary, I don't think you'd be very motivated to work on the slot machine. Well, it turns out some people have already made that decision. Rowan Pandy, previously of Open AI, tweeted the big launch and funding announcement of his periodic labs, saying, Today you will be presented two visions of humanity's future with AI. If you don't want to build the infinite AI TikTok slot machine, but want to develop AI that accelerates fundamental science, raising civilization to Cartyshev One and beyond, come join us at periodic labs.

In another tweet he even added, For what it's worth, I think video models are really cool and could revolutionize robotics arts and education. But this specific trend of distributing them via short form content mobile apps, does make me nervous. I hope my friends at other labs continue the hard work of deploying AI responsibly. Now we're already running super long, so we're not going to get into periodic labs today, but Prin sums it up like this. Periodic Labs has just come out of stealth with the mission to, quote, build AI scientists and the autonomous laboratories for them to operate. At periodic labs, we're building the highest value RL environment, an autonomous laboratory for matter.

For example, to discover a high-temps superconductor, we run RL, rewarding higher critical temperature. In each iteration, the LLM proposes novel materials, which are synthesized and analyzed by robots in the lab. Verifiability makes physics as hill-climable as math or coding. This, too, talent deciding to work on other problems, is a form of market expression and market response to what people perceive as the negatives of using AI for the very questionable social media short-form addiction use case.

Now, I should note that not everyone is as skeptical as the voices that you've heard when it comes to SOAR's potential, while the majority of people that I pulled on Twitter felt that this was an infinite slot machine more than a vehicle for creative expression, it was sort of the opposite on LinkedIn, where 63% of people that responded to a poll of mine said that they thought that SORA2 would represent a creativity explosion, as opposed to just 37% who said brain rot machine. If that says anything, it suggests that for many Open AI has earned the benefit of the doubt and the chance to make a real run at this.

I do think ultimately that AI is going to be the place that culminates many of the conversations that we haven't been having enough about the relationship with technology that we want going forward. And frankly, the ones coming down the pipeline about things like UBI, employment, the very nature of the social contract, are even bigger than those about short form video. So we might as well get our reps in now. And in the meantime, even if you have questions, if you happen to get an invite to this thing, which is extremely restricted right now because presumably of compute constraints, go play around with it. There is something incredibly fun and very clearly powerful about this. The question is whether we collectively can just drive it to the good.

For now, that's going to do it for today's AI Daily Brief. Appreciate you listening or watching as always, and until next time, peace.