Transcription
This has been by far the biggest AI news week of 2025. Like, no questions. In fact, this is the third video this week where OpenAI is taking center stage. And I can't start this video off without first talking about Chat GPT5.
This is the newest, most state-of-the-art model that blows away every single benchmark that's out there. It's amazing at coding. It's better at writing. It's faster. You can make it think for longer or shorter. It's definitely the best version of Chat GPT we've gotten so far and most likely the best model available, period.
Now, I did an entire breakdown on August 7th, the day that GPT5 came out and broke down everything they announced in their keynote. Did a little bit of testing with it and shared my final thoughts about the keynote and the launch. So, I'm not going to go too deep, but I will give you the quick TLDDR of everything they announced around GBT5.
They claim this new model has EHD level intelligence on pretty much every topic. It pretty much crushes every single benchmark with the competition math actually scoring 100% when GPT5 had access to tools. Beats out all the other models on math, science questions, humanity's last exam, and it's currently the best coding model. It also supposedly hallucinates a heck of a lot less and is less likely to deceive you, which is, you know, important.
One thing you'll notice about the new model is that, well, it's less confusing. If you remember, the old model had this drop down with all sorts of options of models to choose from. Well, GPT5 doesn't make you think about which model to pick. It's just got GPT5 and it has some other models here, but I believe these are only on the pro plan. So, for the most part, you're just seeing GPT5 and no other models to use. It decides based on your prompt whether it should think more, think less, and how to attack the prompt that you gave it.
It's also seemingly really fast. You can give it a prompt, and it responds fairly quickly, and it's being made available for free users, plus users, pro users. Everybody's getting access to this new GPT5 model with the higher tier just getting more prompts and more access to it. GPT5 is starting to roll out today, which is August 7th. So, by the time you're seeing this, it should be available to you to All Plus, Pro, Team, and free users, and if you're enterprise or an EDU customer, it's coming in a week.
You could custom tailor the personality to make it more sarcastic, more professional. There's going to be four initial options, a cynic, a robot, a listener, and a nerd. And it's also available immediately for developers inside the API which means that right away as of today you can access it inside of Windsurf Cursor Bolt any of the AI coding tools it's available in. And if for whatever reason you still don't have this model rolled out into your account. You're still seeing the old version. It's already been rolled out into Microsoft's Copilot. So you can use it right now inside of Copilot. In fact, if you head on over to copilot.microsoft.com, you get the chat here and under the dropdown, you can see that I've already got GPT5 available as an option. So, you can chat with it today using Copilot if it's not in your account for whatever reason.
I actually got it to make this vampire survivors clone for me with just one prompt, and it's actually really good. I've got this character that automatically shoots towards the enemies. The enemies drop XP. I can collect the XP. More and more waves of enemies keep on coming as I keep playing, including harder enemies that look slightly different. And it even added in these levelups that work really well. I can click on this one and you can see it added in a new weapon for my character to use against the bad guys. Like one shot, one prompt and it built this. It's really, really impressive when it comes to coding and is now the new state-of-the-art coding model.
Again, I did an entire breakdown of everything, a super deep dive into GPT5 on the day that it was announced. So, if you want to go even deeper into what GPT5 can do, make sure you check out that video. But, you should have access to chat GPT5 right now with these kinds of capabilities. And it's pretty mind-blowing.
It seems to me that all of these large language model companies, Google, Anthropic, OpenAI, they've all realized that the sort of killer use case for these large language models is coding because a you can quickly make these bespoke apps just when you need them for the exact thing that you need them for. But b, if it can't figure out how to solve a problem with the training that's already within its vector database, well, it can write a script and use code to then solve the problem. And it seems like all of the companies are going in this direction because again, if the next word prediction doesn't solve the problem for you, well, creating a little app that then chat GPT runs on your behalf could theoretically solve the problem that the next word prediction couldn't solve, opening up all sorts of possibilities as these things get better and better at coding.
And that was just one of two big releases that OpenAI had this week. They also introduced GPTOSS, which is their open weight models. They released two models, GPTO OSS 120B and GPTOSS 20B. These are models that can actually be run locally or in the cloud and they share no data back to OpenAI or the company you're using them with because they're these open weights locally run models. And they're really powerful. They're about as powerful as the last generation of models that we got before GPT5.
They released this a couple days early and said, "Look, this is an openweight model. Do whatever you want with it. go build off it, iterate it, fine-tune it, do whatever you want with this model. It's as good as all of the models we've put out so far." And then two days later, they released GPT5, which is now the state-of-the-art best model. But this GPT OSS is like as good as the best models before GPT5, just made available open weights.
So, just to sort of solidify this point and beat this dead horse a little bit more, this is a model that OpenAI released for you use completely free. You don't need internet to use it once you have it downloaded. It doesn't send any data anywhere because you could run it offline and locally. And it's as good as the state-of-the-art models up until earlier this week. And this is one where I also made an entire breakdown video on the day it came out, including a complete tutorial on how you can download it and run it locally using LM Studio. So, if you want to know more about this model, what it's capable of, and actually learn how to install it and run it yourself, 100% free, make sure you check out the video I put out all about GPTO OSS from earlier this week.
I was recently accused of being an OpenAI shill, but to be honest, I go where the news is and I talk about where the biggest most important news is happening. And this week, it has been completely OpenAI. And I mean, in lesser OpenAI news, I should probably mention this. They put out this blog post called what we're optimizing chat GPT for and they mentioned that they're worried about people's mental health from using chat GPT too much. So you could in the near future actually get a quick popup while using chat GPT that says this just checking in. You've been chatting a while. Is this a good time for a break? And kind of suggest like, dude, chill out. You've been in here for too much. Go touch grass.
We also got a big launch out of Enthropic this week. They released Claude Opus 4.1. Claude Opus 4.1 was the most powerful, best performing AI coding model on the planet for about 24 hours and then of course GPT5 came and totally overshadowed it. Now, I haven't done a whole lot of sideby-side comparison, and I imagine we'll be testing it more, and there'll be a lot of other channels sort of comparing them to see how well 4.1 does against GPT5, especially in coding tasks, but I haven't gone that deep yet. But this new Claude Opus 4.1 model was totally optimized and designed to be a better coding model.
We can see Opus 4.1 scored a 74.5% on the Swebench verified which ranks how good it is at software development. And again, this was the best we've seen out of any coding model. That is, of course, until GPT5 came along and beat it with a 74.9%. When it comes to the actual API, Claude Opus 4.1 is quite a bit more expensive than the new GPT5 as well. $15 per million input tokens and $75 per million output tokens. G GPT5 in comparison is $1.25 per million input tokens and $10 per million output tokens. So I mean not even in the same ballpark even though GPT5 is slightly better now.
But if I head on over to claude.ai, I can actually access the Cloud 4.1 Opus model as one of my options here. So, let's go ahead and select that and let's give it a quick test prompt and see how it codes something up for us real quick. I'm going to give it this prompt here. Go ahead and pause it if you want to see the full prompt, but essentially I'm asking it to make a fitness app that guides me through a 30 minute workout and targets different muscle groups and has countdown timers and things like that. Let's see if it can get all of this into a single app here.
I will say Claude 4.1 while it's actually writing the code feels slightly faster than GPT5 to me, but I haven't put them sideby-side yet. And we can see here's what it spit out. We got this personal trainer 30 minute daily workout. Key features, seven different 30inut workout routines. Each targets different muscle groups. They're all body weight only. No equipment needed. Smart timers, audio feedback, progress tracking, achievement system, motivation, modern design. I mean, yeah, it did a pretty good job. Let let's make this a little bit wider here and click start workout. Okay, so we got audio feedback. We are getting some sound. Uh we are getting an exercise timer. I could pause my timer, resume my timer. It's wanting me to do squats followed by jump squats. I want to see what happens when the timer runs down here. So once it hits zero, telling me to take a 20 second break. Once this countdown gets towards the end here, now it's going to the jump squats. And we can see next up is going to be lunges. So it's working me through this workout. I'm guessing if I did it on a different day, it would give me a different set of workouts. Pretty damn good for one prompt if I'm being honest.
Just out of curiosity, I tossed the identical prompt into chat GBT5. It wrote a heck of a lot less code. This is a way smaller code file here than what Claude wrote, but let's run the code and see what happens. All right, much simpler website out of GPT5. Daily 30 minute workout. Press start. Let's see what happens. All right. So, jumping jacks with a 30-se secondond countdown. And when the countdown runs down to zero here, we get a little rest. 10 seconds it looks like. And then moves on to push-ups. No audible feedback. We're not getting the sounds. We don't have the sort of tracking and the visual dashboard that we got. So, Opus actually did a better job on one prompt out of the gate with this. So, I think a lot of people are going to debate whether they prefer GPT5 versus Opus 4.1 when it comes to coding. They both seem to work really, really well. Opus 4.1 did this one better on the first prompt, though.
As someone who's not loyal at all to any one model, I think it's going to be really fun to sort of jump back and forth. And if GPT5 gets hung up on a coding problem that it can't seem to solve, well, jump over to Opus 4.1 and see if it will solve it. It'll be a lot more expensive with their API, but maybe it'll solve a problem that GPT5 couldn't. And vice versa, and you could switch back and forth to whatever one works the best. And if you want a really inexpensive coding model, you use Gemini 2.5, which is probably the best economical pricing model. I just love the fact that we have all of these options, and all of them are so good.
Now, we also got some news out of Nvidia recently. They launched their version 4.0 with the new AI Twin feature. This feature allows you to upload videos of yourself and create a digital clone of yourself or use one of the pre-created avatars that they've already got. So, check this out. If I log into my Nvidia account here and I click on create AI video, there's a new button here that says AI twins. I have the option to create a twin from YouTube, record fresh, or clone just my voice. So, for this video, let's record from scratch and create a fresh new avatar. I'll click on recording to avatar. It tells me to upload at least 60 seconds of video and to make sure I give Invidia AI permission with my words in that video. So, I'm going to go ahead and record that real quick and then we'll go ahead and upload it and see what happens.
And now that I have my AI avatar, I can go in and generate a video. So, I can give it a prompt like three grrowy facts that will blow your mind in 30 seconds. Select my AI twin here. We'll click continue. And now I can generate my video. Think gravity is pulling you down right now? Plot twist. You're actually accelerating upward through spaceime. Here's what else will blow your mind about gravity. If you fell into a black hole, time would slow down so much that billions of years would pass in the outside universe before you even cross the event horizon. Honestly, Nvidia's pretty great for lastminute re-shoots, creating simple and fun ads, or just messing around cuz sometimes that's fun, too. And you can actually try this AI avatar feature for free over at invido.io, I will make sure it is linked up in the description. And thank you so much to Nvidia for sponsoring today's episode. Now, let's get back to it.
And if just the few things I've already mentioned weren't enough to make this like the biggest week ever, we got news out of Google DeepMind as well. They showed off their new Genie 3, which is a front tier world model. And the Genie 3 website has all sorts of other demos. Like this is navigating maybe the surface of the moon uh or navigating the surface of a volcano. Here's one where somebody's like on a jet ski during a festival of lights navigating. And you can actually see it just hit that light and the light bumped off to the side as it hit it. Walking on a pavement in Florida next to a two-lane road. And there's waves crashing over the side here. And the user is able to look around in real time as this is happening, tracking a jellyfish underwater, flying a helicopter around over a coastal cliff. They're actually controlling this helicopter and where it's pointing using their arrow keys. This is just wild. And there's a whole bunch of other demos on their website that you can take a peek at to see what this is really really capable of because this this is honestly just absolutely wild.
Now, we don't have access to actually play with this ourselves yet. They're making it available to early testers. I might be able to get my hands on an early test to play around with it, but this is absolutely wild that this is all being generated. Every single frame you're seeing is being generated as the character, the player uses the arrow keys to move around on the level. Like, they even jumped into this like flying island. Wild stuff. A lot more demos here. I won't spoil them. Check out the Genie site. All links will be available in the description. But mind blown what you can do with this stuff already.
Just as somebody who's followed along to AI for the last fiveish years, I am just so blown away by what people are already taking for granted. Like 5 years ago, none of this was even close to possible. I don't even think people were thinking about making this kind of stuff 5 years ago. And here we are already better than I can even imagine. Jacob Bower here on X showing another use case where you can actually give it a video. This was a video generated with VO3 and halfway through the video it switched to Genie 3 and now he's controlling the flight. So you can actually feed it videos and then take control of the video. I mean, are we not entertained?
Google also rolled out a feature this week that allows you to create personal illustrated story books using the Gemini app. And I haven't tried it yet. So let's see what it does. Try it today in the Gemini app. So when I click the link, we can see that it's on storybook here. Let's see how creative it makes the story book if I actually don't give it much details. Create a story book with custom art that tells a story about a young boy going on an adventure. Everything about the adventure is up to you, but should be original and unique. Let's see what story it generates for us, just not giving it much to work with.
So, it immediately follows up with some more details. Reader's age, let's go ahead and say 10 years old. Illustration style, let's do cartoon. And then tone. Let's do humorous. And let it do its thing. It took maybe a minute and a half, but we've got our book here called Sam's Epic Sock Adventure. So, we've got our cover here. I guess that's Sam. Pretty decent art. Sam, a boy with a map in one hand and a flashlight in the other. Stood at the top of the basement stairs. Today, Sebastian, he declared to his unimpressed cat. We venture into the treacherous laundry room cavern to find the lost socks of legend. Looks like the whole thing's 10 pages. Here's page two. I'm not going to read the whole thing, but feel free to pause on any page if you feel like reading the whole book. Page three here. I mean, the art is all very consistent as well, which is pretty impressive. The character looks the same. What's going on with that dry air over there? Page six, page seven, page eight, page nine, and finally, the kid gets his socks. And we've got a button that says start over. I mean, pretty fun little app. I mean, it would be cool to make little books for your kids. I'm curious if it can make more of like an adult book, too. Not what you're thinking. You know, the book Go the F to Sleep that people give new parents is like a funny gift book. I wonder if it can generate that kind of thing.
So, I'm going to go back to our little story book gem here, and I am going to censor this a little bit. This is a family-friendly channel, but create a story book called Put Your Away. That's about a father who is constantly picking up after his children. Is it going to make it for us? Let's find out. One thing to note is this time it didn't ask follow-up instructions. It seemed to know its task here. I'm actually kind of shocked that it did it. We've got our book here and our cover with our title. Sorry, I'm censoring it, but you know what it's about. Dad was a patient man, a kind man, a man who loved his children more than anything. But if he had to pick up one more goddamn toy, he was going to lose his mind. All right. All looking good. Similar art style. Page four. I mean, the art's all really good. It all has a very consistent design. Everything's looking great here. Page seven, can find their toy. Page eight, Leo and Mia looked at the mess. Then they looked at the empty toy bins. Then they looked at each other. The tiny light bulb went on. Slowly, they started to put their away. I mean, for the most part, other than the title, and then maybe two swear words in the entire book. It's fairly friendly. It wasn't quite as uh swear as the go the F to sleep book, but I'm actually shocked that it even allowed me to do this in the first place. I thought it was going to get denied.
Grock rolled out a new feature called Grock Imagine which can create images and video. And when it does create video, the video even has audio similar to VO3. And it's really, really fast. Now, at the moment, you have to use it inside of the Gro app. And it does say here it's now live to all Super Grock and Premium Plus subscribers. on the paid plans. However, Elon Musk took the X to say for the next few days, Grock Imagine Video Generation is free to all US users. Download the Gro app to try it out. So, you can actually use it for free on the Gro app if you're in the US.
Now, what if you're not in the US? Well, somebody went to X to say, "Bro, just make it free for all users worldwide." And well, Elon said, "Use a VPN." So, apparently, if you're using a VPN and you're outside the US, you can make Grock Imagine work. So, let's take a peek at it on the iPhone here. So, inside the Gro app, you can see at the very top there's an imagine button. Also, down at the bottom there's a create videos button. Let's just go ahead and click on imagine at the top. And you can see that there's a whole bunch of images here that other people have generated. And one of the easiest things you can do to animate something right out of the gate is just to pick an image that's already been created. And you can see there's a button that says make video. Now, make sure when you use this make video that uh you pick an option. There's normal, fun, spicy, and custom. And spicy mode, well, it's not safe for work.
Here's what happens when I just do like fun mode on this Lego Mona Lisa here. It starts to sort of animate. There is audio, but it's not playing through my computer, so I'm holding my mic up. Hopefully, you can hear that. It's a little weird. It did, however, make three other generations that you can pick between to see which one you like, but they're all kind of very similar and weird with this specific video. Most of the videos I've done don't come out this funky. If I open up this mermaid one that somebody made. Let's see what happens if I make it spicy. I'll probably have to censor this. So, that wasn't too bad. That wasn't as spicy as some of the other videos I made. You can see though that I've done some of my normal tests. Monkey on roller skates, Wolf Howling at the Moon, it did pretty decent. But the first time I ever did spicy mode out of curiosity, it did this, which I don't even think I can show you. So yeah, it will it will do what you think it will coming from an app made by Elon Musk.
We got a new image model called Quinn image this week, which a lot of people are claiming is the best image model out there right now. It can do text. It can do realistic images. It can do cartoon images. People are just raving about this. You can play with it by going over to chat.quinn.ai and clicking on the image generation model here. And I can generate an image like a wolf howling at the moon with some text seen on the moon. And it pretty much nails it. And I'm not even logged in over here on Quinn. This let me just generate it for free on their website. I'm sure I'll run out of generations at some point and they'll want me to log in. But as of right now, I generated this image not even needing to be logged in. And the text is all pretty good now. I mean, compared to where we were a year and a half ago with getting texted images, it's crazy. It's just like a given now that an image model will be able to do good text.
Speaking of image models, there's a new model out of Leonardo, a company that I do advise for. So, I want to just disclaim that real quick, but it's probably my favorite model of the moment as when it comes to like models that appeal to my visual aesthetic sense. Like, I really like this new Lucid Origin model and the images that it makes. It reminds me of what we were getting out of MidJourney back when Midjourney V4 came out and everybody was really really impressed. Like I really like the aesthetics that we're getting out of this. So if I'm in my Leonardo account and I click on image here up on my models, I can select Lucid Origin here. I'm going to give it a prompt of a vampire fighting a werewolf. Let's have it improve my prompt for me real quick. And let's generate that. And we can see it made a very sort of painterly looking image for each of these, which isn't bad. That's not really the aesthetic that I like to go for. So, let's generate a new variation of the prompt here. And we get some very cool like comicbooky looking art that I really like the aesthetic of. Here's some more that I generated. This one almost looks like a game cutscene or something. This one you got like some realism going on. So, another really good image model.
And another feature that Leonardo just rolled out is with V3. If I click into the video mode, now you can actually use VO3 and start with an image for your video. This is brand new. You weren't able to do a starting frame. Before it was only text to video. Now you can do image to video. So I can click here to add a start frame. And I can pick from images that I've already uploaded here or I can pick from my generations. So let's pick one of these werewolf generations. Confirm it. give it a basic prompt of a vampire fighting a werewolf and let's see what we get out of it. And here's what it generated. I mean, it kind of feels like a fight in slow motion and the vampire is teasing the werewolf to eat his arm. But it's really cool that Leonardo has image to video generation now.
And yes, I'm still going with more crazy releases this week. 11 Labs released a new feature where now you can generate music from 11 Labs. you were already able to do text to audio, uh, existing speech to a different voice, sound effects, drum machines. Now they're competing with SNO and Udio. And you can generate full-on music directly inside of 11 Labs. So, let's jump into 11 Labs and try this oursel. I have not generated music, so this is going to be my first time hearing it.
Inside 11 Labs, I can see in the left sidebar, we have music. And right in the homepage, we also have music. So, let's click into the music section. And it says, "Describe your song." Um, we can also describe the duration. We'll leave it on auto. It looks like it will generate two variations of our song. I'll give it the prompted early 2000s pop punk anthem with three-part harmonies about how you should watch Matt Wolf on YouTube to keep looped in on the latest AI news. If you've watched my channel for a while now, you'll know that I'm a fan of early 2000's pop punk music. And when I've tested Suno and Yo and tools like that in the past, that's the style I generate because I want to see how they compare. So, let's see what it does. It's interesting because it generated all the lyrics right away. And then down at the bottom, we can see like a timeline. And as we watch it, you'll actually see new audio get generated down here in the timeline. So, verse three hasn't been generated yet, but it's generated all the way up through the chorus. Just kind of cool to watch this generate in real time in front of you.
All right, so let's take a quick listen to the song. He's the ultimate scoop. Crank the volume up. Get your brain on track. Join the re. >> Not bad at all. Now, personally, I still think maybe Suno and Yuio do a slightly better job, especially with the vocals. You can still kind of hear that the vocals are AI and the words sound a little bit weird, like they blend words together a little too much, but it sounds good. I mean, I was bobbing my head to the song. I feel like Sunno still makes catchier songs out of the gate, but again, this was the first prompt, so there's probably a lot more I can do to really fine-tune this and dial it in. I will play with it a lot more. Out of the gate, though, I don't necessarily think it beats Suno, but it's it's really good and it's really fun to watch it generate in real time.
I do want to do one more quick test. I'm going to click on music again and I want to try to generate something without vocals because you can actually create like stems like just a baseline or just a guitar rhythm or just vocals without background music things like that. So let's try a blues rhythm with just rhythm guitar bass and drums, no vocals and no lead guitar. It should be in the key of G. Let's see what it generates for us. is the idea being you can generate like a rhythm backing track. If I get something that's guitar, bass, and drums, no singing, no lead guitar. Well, then I can like jam over the top of it with my real guitar if I wanted. We can see up at the top include styles, blues, rhythmic groove, instrumental, shuffle, key of G, exclude styles, vocals, lead guitar, solos, harmonica, keyboards. And this is what it did. Okay, one second. I got to tune it real quick. [Music] All right. So, super super fun, especially if you're a musician and you want to, you know, create other little parts of the song and then overlay your own instrumental or jam to a song that AI made with you. Super super fun. Really excited about that. And that was the last demo that I wanted to share today.
There are a few other little super quick things I wanted to share. So, let's jump into a rapid fire section. If you're a big fan of using MidJourney, they just rolled out an HD video mode if you're one of the pro or mega subscribers. So, you got to be on one of their higherend paid plans. It is a little bit more expensive to generate, and by a little bit, quite a bit more. 3.2x more expensive, but you get four times as many pixels than the default SD video. So, if uh you need higherend video coming out of MidJourney, well, it's now available if you're on one of those plans.
The Google company Kaggle is basically making a bunch of games that they're going to use to test different AI models against each other. And the first thing they're doing is an AI chess tournament to evaluate which models are currently the best at chess. It's a three-day long AI chess battle, and it's the first in a series of tournaments hosted by Google to find out which model's the best at games. The models will compete against each other in a range of strategic games designed to evaluate their thinking and reasoning capabilities. For the first round, they've partnered with chess.com, so pretty soon we'll know which of the available AI models is best at chess. And then after chess, they're going to put it up against a bunch of other games. So, should be fun to watch and follow along to.
And finally, Elon Musk went to X to say that Gro 5 will be out before the end of the year and will be crushingly good. And he made this post on the same day that GPT5 launched. And if I'm going to be honest, it feels a little bit like there was a big news out of OpenAI this week, big news out of Google this week, uh fairly decentsized news out of Anthropic this week, and X was a little bit left out. Now, they did put out imagine, but it doesn't feel like as many people are talking about it. And so Elon kind of went, well, Grock 5 is coming and it's going to be really, really good. That's kind of how I feel about this tweet here. It's probably not really the motives. I could be reading way too much into this, but supposedly we'll be getting a Grock 5. And if you remember, Grock 4 just came out like a month ago. So crazy that he's already talking about Gro 5. Also, we have to keep in mind this is Elon. Elon's not always the best with timelines. So, if I had to guess whether or not Grok 5 will actually see the light of day before the end of 2025, my guess is probably not, just based on the track record. But, I'm hoping for it.
I love seeing new models. I love testing new models. I'm trying to find new ways to test new models. My goal is to figure out some sort of benchmark of like how is this actually useful for everyday people instead of the crazy math and science benchmarks we're getting now. I want to develop my own. I haven't sort of landed on exactly what that looks like yet, but I love all these new AI models coming out so that I can sort of test them against each other. So hopefully we do get Grock 5 by the end of the year. I'm just not going to hold my breath.
And that's what I got for you. I know, big week. Crazy week. So much stuff happened. So many new releases. GPT5, GPT OSS, Enthropic Opus 4.1, Genie 3, 11 Labs Music, new models in Leonardo, new Quinn image models. It has been a week, a crazy week, a fun week. If you're paying attention to AI, this has been one of the best weeks ever. So much good stuff and it just feels like it's accelerating even more. And as you know, I'm here for it. So, if you want to stay looped in on the latest AI news and get tutorials on how to use all of the latest cool AI stuff that comes out, make sure you like this video, subscribe to this channel, and I will make sure more cool stuff shows up in your YouTube feed all about the latest cool technology that you can play with right now. But that's what I got for you for this one. Thank you so much for hanging out with me and nerding out with me. Amazing week. Hopefully we get more exciting weeks like this this year and I'm here for it. Hopefully you'll join me for future episodes like this. And uh yeah, that's all I got for you. Thanks for nerding out and hopefully I'll see you in the next one. Bye-bye.
Thank you so much for nerding out with me today. If you like videos like this, make sure to give it a thumbs up and subscribe to this channel. I'll make sure more videos like this show up in your YouTube feed. And if you haven't already, check out futuretools.io io where I share all the coolest AI tools and all the latest AI news. And there's an awesome free newsletter. Thanks again. Really appreciate you.