AI News: Here's What We'll Get From AI in 2025

Well, Happy New Year. Surprisingly, there hasn't been a ton of AI announcements between Christmas and New Year's. Okay, maybe that's not that surprising, but there has been a few. I'm going to quickly break down the few small announcements that happened over the last few days, but then I'm going to spend more time on predictions of what I believe is going to happen with AI in 2025 as well as what you think is going to happen in 2025. I asked you guys over on X what your predictions for AI and other emerging tech was in 2025 and received one responses. So we'll dig in a little bit on both what I think is coming in 2025 and what you guys think is coming in 2025. But first, here's a quick rundown of the news that happened in the last few days. On December 27, OpenAI pretty much confirmed what everybody saw coming and that they are going to attempt to convert from a nonprofit to a for profit company. They believe that's the best way to make the company more sustainable and actually achieve their mission. They plan to transform the existing for profit Delaware Public Benefit Corporation, which requires the company to balance shareholder interest, stakeholder interest, and public benefit interest in the decision making. The nonprofit would essentially become an investor in the for profit. They will hold a large amount of equity in the new for profit. Now this whole conversion of a nonprofit to a for profit has actually been met with quite a bit of resistance from outside companies. Elon Musk has tried to block this from actually happening and now the godfather of AI, Geoffrey Hinton, has also said that he does not think it's in the best interest for OpenAI to convert to a for profit. OpenAI was founded as an explicitly safety focused nonprofit and made a variety of safety related promises in its charter. It received numerous tax and other benefits from its nonprofit status. Allowing it to tear all of that up when it becomes inconvenient sends a very bad message to other actors in the ecosystem. So although OpenAI wants to convert to a for profit, there is still some opposition out there that might try to block that from happening. I've mentioned in one of my recent live streams that I personally don't find this that exciting of information because it doesn't really impact how I'm currently using AI on a day to day basis, which is really the main focus of my channel is how people can use this stuff on a day to day basis. But this is hot and in the AI news right now, so I figured I would share this stuff quickly. Also, I've seen a lot of the tech websites point this out, so I'll point it out really quickly as well. OpenAI failed to deliver on the opt out tool that it promised by 2025. If you remember back in May, OpenAI said it was developing something called the Media Manager, which was a tool that would allow content creators and websites to opt out of OpenAI training on any of its data. They said they intended to have this launched by 2025. Well, we are officially in 2025 and it hasn't even launched. And supposedly according to people within OpenAI, that feature was rarely viewed as an important.

Someone who used to work at OpenAI said, I don't think it was a priority. To be honest, I don't remember anyone working on it. So who knows if it's even something that will ever get created. But as of right now, we haven't heard any more about it since back in May. Google, who ended 2024 with a bang, showing off VO2, the Willow Quantum chip, their Genie 2 model, new versions of Gemini, the Notebook LM. It just seemed like everything kind of came together for Google at the end of 2024 and they seem to be just getting started. Google's CEO told employees to gear up for a big 2025. He told employees, I think 2025 will be critical. I think it's really important we internalize the urgency of this moment and need to move faster. As a company, the stakes are high. These are disruptive moments. In 2025, we need to be relentlessly focused on unlocking the benefits of this technology and solve real user problems. Google sort of had a reputation during 2023 and a lot of 2024 of not shipping as fast as companies like Anthropic and OpenAI when it comes to releasing AI, they would show off features but then not roll them out. But towards the end of 2025 they really flipped that switch and started actually releasing a lot of really, really cool stuff. So Google is one of the companies that I'm pretty excited to see what they produce in 2025. With the current state of AI technology, there better time to get into creating online video content. So for today's video, I partnered with Opus Clip because it's been one of my secret weapons to turn long form video content into tons of short form content. If you've been watching my channel recently, you've probably noticed I've been doing live streams lately. Well, Opus Clip has allowed me to turn those live streams into hundreds of short form clips. That I'm able to then publish on YouTube and Instagram for thousands and thousands of extra views on my videos. You feed it one long form video, it will analyze that video and find the clips that are most likely to go viral as short form clips. It also reframes the entire video. All of these were shot in a 16 by 9 wide format video. Opus Clip automatically resized it to 9 by 16, put my face on the videos and whatever I was showing off on the clip on the video as well. And they keep adding new features as well. They've got an AI voiceover feature, a speech enhancer feature, a feature to automatically add B roll, and their editor is really powerful. I can use AI to enhance the SPE speech, remove filler words, remove pauses, even add sensors to my videos. If I don't like how the clip was laid out, I can easily rearrange in the editor, put my head on the top of the video and the thing that I'm showing on the bottom of the video. And just like that, you're seeing the secret to how I've managed to produce so many short form videos off of one weekly live stream. Opus Clip has made it so easy. So if you're thinking about getting into video content creation, 2025 is the year to do it. Tools like Opus Clip make it extremely easy to create loads of valuable content. And what you're seeing on my various channels is proof of that. You can learn more about Opus Clip in the link in the description. And thank you so much to Opus Clip for sponsoring this video. Moving over to Meta News real quick. This is something that I've heard quite a bit about that I'm not super excited about from Meta. Apparently Meta wants to release a whole bunch of AI fake people on platforms like Facebook and Instagram to essentially provide people that get very little engagement with more engagement to make their platforms more sticky. And I personally think this is a very weird, like almost dystopian move by Meta. Now, Meta's made a lot of good decisions and I love a lot of what Meta did, but this is one where I'm kind of stepping back and going, is this really where you want to take your platforms? The tech giant is planning to roll out millions of generative AI characters that will appear as users on Facebook and Instagram. So they're going to make them look like real people on these platforms. We expect these AIs to actually over time exist on our platforms kind of in the same way that accounts do. They'll have bios and profile pictures and be able to generate and share content powered by AI on the platform. That's where we see all of this going. And this came directly from Connor Hayes, the Vice President of Product for Generative AI at Meta. This article says that while bots are typically viewed as a negative aspect of navigating social media platforms, Meta's decision to purposefully introduce non human users into its social media ecosystem is linked to the company's desire to appeal to a younger generation of users. So while most of these social media platforms are doing everything they can to fight bots and eliminate bots, Meta's idea of the future of social media is to introduce a ton of new bots and let humans interact with bots, likely in a way where they're not even realizing they're interacting with bots, to increase the stickiness and retention of users on these social media platforms. I don't like that. I personally am not a fan of where that is headed. If I'm on social media, I want to know I'm interacting with other humans. I'm sure there's going to be some sort of disclaimer that's noting that they're talking to an AI, but their goal is definitely to create a social media platform where people feel like they're getting more engagement, even when that engagement isn't actually coming from real humans. This week, Hugging Face introduced small agents. This is a simple library that unlocks agentic capabilities for language models. Now, I haven't dug into this myself too much, but this appears to be a suite of tools available on Hugging Face that allows anybody to create agents with existing tools that are already out there. But it is something that I will likely play around with in the future, and once I have a better understanding of it, I'll share more thoughts on it. Nvidia this week closed an acquisition on a software company called Run AI. It's estimated that it cost Nvidia about $700 million to acquire Run AI. And if you're curious what Run AI is, it appears to be a platform that helps optimize AI infrastructure. We can see in the press release here it says in addition, we plan to open source the Run AI software to help the build better AI faster. While Run AI currently supports only Nvidia GPUs, open sourcing the software will enable it to extend its availability to the entire AI ecosystem. And in the last little bit of AI news that I'll share here, there's a new Nintendo patent that appears to reveal Switch 2, and the patent claims that it's going to use deep learning super sampling. Basically, this is technology that uses AI to upscale the resolution of games, effectively allowing developers to achieve higher graphical settings and better frame rates on weaker heart hardware. So essentially what this is saying is that the next version of Nintendo Switch, they don't have to make games that actually are in 4K. They can make the games in like 1080p or lower resolution, ship the games in that lower resolution, then when you play the game on the device itself, it's going to upscale back up to like a 4K resolution and give you that higher quality video. This is likely going to be the future of a lot of game systems where they can make smaller and smaller file sizes by downscaling the resolution of the game. And then while you're actually playing the game, the console or the computer or whatever you're using to play the game then re upscales it back up into that like 4K, really, really high definition video. And that's pretty much all the most interesting AI news that I found for you this week. Some of it's not even that interesting, but I did want to share what little news did trickle out this week. But now let's quickly talk about AI predictions for 2025. My number one prediction is that 2025 is the year of AI agents. We're going to hear AI agents everywhere. We're going to hear the word agentic a lot. So much that you're going to get sick of hearing the word agent. Because I think everybody is going to build AI agents into their platforms. And what I mean by AI agents is using AI that can also use tools and do things on your behalf. Imagine using Siri to look up a flight for you, help you book it and then add it to your calendar and it would do all of those for you. Or telling your Google Assistant to help you schedule a meeting with somebody and your Google Assistant goes and sends that email with somebody, goes and does some back and forth with emails to confirm a time and once the time is confirmed, adds it to your calendar, things like that. I think we're going to see a lot more of in 2025 where you're using some sort of like voice assistant and it goes and takes actions on your behalf. We know that Anthropic got huge investments from Amazon this year, so your Amazon Alexa is probably going to have much more of this kind of technology built into it. We've seen what Google is trying to do with Gemini in their Project Astra. We've seen what Apple's doing by baking in AI into Siri and connecting it to ChatGPT. We've seen what Meta is doing by rolling AI into all of their messaging platforms. So I really, really think all of them want this sort of like agent tool use kind of functionality. Claude even put out their tool use feature. Now it's not really easy to set up for people yet. It's like a very, very early testing sort of scenario. But they've already shown it off and I've already played with Claude that can use tools. OpenAI said they're expecting tool use to roll out in January of 2025 for ChatGPT. And along with that, I think we're going to see more AI baked into everything. When I was at CES last year, they had AI in refrigerators, AI in laundry machines, AI in toilets, AI in ovens, AI in barbecues, AI pillows, AI beds. Literally everything had AI. I'm going to ces, like I mentioned next week, I'm expecting even more of that. Like every booth is going to have some sort of G, but now with AI in it, I really think we're gonna see that. I also think we're gonna see a lot more of the AI wearables. Like we're gonna see more of the sunglasses and the earbuds and the wristbands and things like that that have AI baked into it. The Meta Ray Bans were a big hit. They continue to sell well and do well. And Meta's already said that in 2025 they think they're gonna add some sort of like heads up display in them so that you can actually see text and subtitles and things like that right in your eyes inside of your sunglasses. We already saw that. Project Astra is gonna get like a glasses version soon, so we're going to see more of those wearables. We've also seen sneak peeks of like earbuds that you can wear that will do automatic translation for you and things like that. So wearables I think are going to be a really, really big thing in 2025 that we'll hear a lot of talk about. I mentioned video models getting way better from 2023 to 2024, and I think we're going to see that trend continue. Sam Altman from OpenAI said that there's a lot of plans in the works for improving Sora. Right now we have this Sora Turbo model. We don't even have the full Sora model yet. So we're going to see big leaps there. We from Google, which most people still don't have access to, but it's really, really good. We've got the Chinese models like Cling and Minimax that are really, really good. You've got Pika and Runway, which are both constantly innovating in their own ways. So we're going to see video just get better and better and better in 2025. I imagine we're going to get videos that are longer, you know, generate up to a minute or longer with our AI videos, they're going to get higher quality. The physics understanding is going to get a lot better where they understand which leg is the, you know, the leg that's furthest from the camera and which leg is the closest to the camera and understanding environments a little bit better. I really think the AI video is going to get really, really good in 2025. Like it's already pretty good. I think it's going to get to that indistinguishable from is this a real video or is this generated with AI? I think people are going to have trouble determining the difference. I think we're going to see more self driving tech in 2025. We know Tesla's got their Cyber taxi which is coming out in 2026, but we also know there's companies like Wemo who are connected with Uber. There's rumors that Amazon might try to work some sort of acquisition with Lyft and work in autonomous vehicles into Lyft as well. And I think a lot of these companies are going to try to do this stuff before the robo taxis or cyber taxis or whatever they're called are released by Tesla. I think wemo is going to try to get ahead of it. Amazon has a company called called Zoox, I think Z O O X which also makes like autonomous little like buses and vehicles and things. So I think if they end up working with Lyft really closely, we might see Lyft actually have autonomous vehicles inside of their fleet as well. So I think we're going to see a lot of that bubble up even more in 2025. I think we're going to hear a lot more talk about quantum computing. We recently saw that Willow chip from Google which made huge leaps in what quantum computers can do and minimizing the error rates of these quantum computers. IBM has similar chips that they're working on that are going to be released in 2025. And I think the talk around quantum computing is really going to bubble up more and more in 2025. So those are really my predictions for 2025. AI agents, AI baked into everything. More and more wearables like glasses with heads up displays and earbuds that can automatically translate things. The continued progression of video models, they're only going to get better and more impressive in 2025. More and more of the self driving tech and self driving cabs like the wemo, like the Zoox, like the Tesla version, and finally a lot more talk around quantum computing. I'm sure there's quite a bit that I am missing and also my timelines are often very off, but I really think those are the areas we'll see a lot more in 2025. Now I'm going to do a quick run through of some of your predictions that you shared on X with me. I actually haven't most of these yet. Some of them I kind of accidentally saw in my notifications, but most of these I haven't read yet. So Bilavel says AI video shatters the visual Turing test. Totally agree. That's one of the things I predict as well. Jason Moffatt, one of my buddies, he says, I think unless you're way ahead of the curve, many people are going to be penalized using AI because it's creating so much of the same old same old from people who aren't very creative. I agree with that. From like a content marketing standpoint, I think if you're relying on AI a little too heavily to create your content for blog posts and things like that, it's probably not going to do super well. But if you're using AI in other ways, like if you're using AI to create really good videos, really good AI imagery, using it to help like brainstorm ideas, things like that, I still think AI is really, really valuable. I just think AI generated content is also going to create just a flood of AI sort of slop on the Internet. And that I'm not looking forward to. But I also think Google and some of these platforms are going to figure out how to sort of derank that stuff. Right now it's really, really tough for algorithms to tell whether or not something was AI generated. I think that's only going harder. But I also think a lot of these AI generation tools might do some sort of like watermarking thing to help determine whether it was generated with AI or not. We'll see. I don't know. I don't know how that's going to play out. I think SEO in general is just going to be totally overhauled pretty soon as a result of AI. The way we search is changing. We now sort of ask questions and use prompts instead of searching as much. And the way we create content is changing because it's so effortless to create content now. So I just think SEO as a whole is going to change a lot. Johannes Copeland says, we're going to get damn close to AGI with agents that actually work effectively replacing large parts of existing jobs. Couldn't agree more. Nick Silver says, I personally think 2025 is the year of AI agents. I think that a lot of people that are paying attention to like the AI verse see AI agents is the next thing. If you do want to post a blog post on WordPress, it's not going to be go to ChatGPT, have ChatGPT write it for you, you copy and paste it into WordPress and then you publish it. It's going to be more like, hey, go write me an article about X. And then your AI agent goes and does the research, writes the article, copies and pastes it to WordPress for you, generates a thumbnail image for you, and then presses the publish button for you. So all you did was instruct it on what you wanted it to do and it did everything beyond that point. I think that's where like AI agents and content creation are gonna go. Josh Marino because of the Jetsonship, there will be an explosion of entertaining robots like Pokemon and other characters for kids. That would be pretty cool. I'd love to see that. You'll see a ton of video AI produced entertainment in a fraction of the time. It'll allow creators to get their voices out there instead of spending weeks to produce something similar. Zoom says this year will be about implementing AI in the workspace effectively and have it be able to carefully complete tasks by itself. Yeah. So basically like the agent sort of stuff, we need AI that reliably produces results and that doesn't need too much supervising. Agreed. Yeah, that's. That's where I think like the AI agent stuff is really going to happen this year. It will cause a great deal of confusion. Devising ways to discern truth from fiction will be come a huge priority. Totally agree. I think we're already there with AI images. I think people are already pretty good at creating AI images that can fake people out and make them think they're real. I think we're going to get there with video in 2025. In fact, I'm very confident we're going to be there with video in 2025. Raymond L predicts that it will have the first global incident caused by an AI agent going haywire. I hope not but you know there's going to be people in their basements trying to figure out how to wreak havoc with these various AI models. Sir Charles says AI will make its way into some big movies one way or the other. And not only the thing. Yeah, I agree. I mean, I think we're already seeing it. We're already seeing AI help replace faces with like deceased actors and using AI to get like voiceover and things like that. So I really think we're already starting to see it. I think that a lot of Hollywood and a lot of game studios and things like that are going to sort of hide the fact that they're using AI at least in the near term because they know there's some backlash. But I think a lot of these companies are already using it. They're just not talking about it as much. Victor thinks that we'll have the movie heard next year. But using AR glasses instead of earbuds, I could see that. I mean, with Project Astra, I've already started to get glimpses of that. If we get the advanced voice mode with chat GPT but in glasses, I think, you know, you get that kind of scenario. Nux says I want to see an AI that takes video camera and microphone as an input and produces video as an output. Depending on what the input is, the output could be an avatar conversing or a graphical illustration or a movie, whatever makes the most sense for the input video. I would love to see that. I don't know if we're gonna get that in 2025. I'm pretty sure that Google is working on to do that inside of Gemini. I mean, their model is multimodal from the start, so it can understand video. I don't know how well it can produce video yet, but it can understand video. And then when you look at OpenAI, you know, they're trying to add video understanding as well. But they also have a platform that outputs video in Sora. You know, Google has VO2, so both of these platforms have models that can output video. Both of these platforms are able to take in video. You'd think they would get there soon. Whether we see that in 2025 or not, I'm not sure. Self improving AI agents full length movies. AI generated names AI agents AI agent takeover. Greg says AI gets personal. Smart assistants that truly know you Ethics take center stage. Stronger regulation and frameworks to ensure responsible AI development. I'm really hoping for this as well. Creativity goes hybrid. AI becomes a true creative partner revolutionizing how we make art, music and content. Yeah, that one, I think we're already there. I think people that really understand and leverage AI are already kind of using it as a partner, not as like the only tool. Like me personally, I create art in midjourney and Leonardo and Ideogram and all of these various places. And I pull it together in Photoshop and then I throw it into Canva to add some, like, text. Really easy. So I'm already kind of creating my own art by using a ton of different tools and trying to pull it all together. Richard o' Connell says still no one will care about Grok. I actually kind of disagree on that one. You know, Elon is trying to build like the world's largest data center. He's also trying to build an external GROK app that's outside of X. So if he builds this data center and he builds an app outside of x, that's like ChatGPT, but like an uncensored version, I actually think a lot of people will care about Grok at that point. I just think a lot of people don't use it because you have to use X Twitter to actually use Grok right now. And it's not like the best of the large language models. It's not the best image generator anymore. It was when it used Flux, but it's not right now. But I think when it's its own app, if they maintain the fact that they want it to be like this uncensored model that you can do anything with, I think Grok will be very, very popular among people that want to use an uncensored version of a ChatGPT in a standalone app. Tesla full self driving will go unsupervised. Totally agree. I think we're seeing that in 2025. Crypto shrimp. The crypto solutions for proof of personhood need to be developed. I do think this is a decent use case of blockchain technology. So I would love to see something bubble up that does that. Well, AI Compute will slow down a bit, but its applications will become much more embedded and AI will have another Chat GPT moment. I can see that. Restructuring democracy says apps become ideas, only thing necessary, AI takes care of the rest. That's totally the future that I think all of this is working towards. Whether that's 2025 or not, I don't know. But I would love to be able to go to a platform and say, I need an app that does this and it just spits out the app and it works the first time, or I only have to give it one or two more Prompts to add some features. I think we'll get there. I don't think we're too far off from that. Whether we see that in 2025 or not, I don't know. We don't have access to OpenAI's O3 yet. Maybe O3 is capable of doing that kind of thing, but I don't know yet. AI powered automation. Yeah, that's kind of like like the, the agent stuff. Quantum computing breakthroughs. Totally agree. I mentioned that as well. Extended reality is the new norm. XR technologies will become mainstream. I thought that in 2024. I can see it maybe being 2025, but I don't think it's going to be the big headsets. I don't think it's going to be the meta quests or the Apple Vision pros. I think it's going to be more the smaller glasses form factors. I think it's going to be the Ray Ban metas, but with heads up displays. The Astra glasses with the heads up displays built in, they've got to look like something you would wear out normal only for them to get adopted, in my opinion. But I do think we will see more of it in 2025. Like I mentioned, 2025 is the year of agents. Agents galore. Agents going mainstream. Someone's going to do a series of short AI movies that will hold up and become popular probably on YouTube. I can totally see somebody creating like a love, death and robots style show that was all generated with AI and does really, really well if the storytelling and the visuals are both there. I think you need those two elements. Good storytelling that hooks people in and good visuals that don't bore people with the visuals. Glasses with real time streaming functionality. Google glasses, Apple glasses. I don't know if Apple's actually going to do any sort of glasses. Maybe, but I haven't really even heard rumors about Apple glasses. Glasses powered by 01 or 03 definitely see that happening. AI agents everywhere, custom droids, quantum computing, regenerative medicine with AI. I would love to see this happen in 2025. I just, just think that the sort of legal roadblocks really, really slow this portion down because I thought this in 2024 and nothing really manifested as fast as I thought it would. But these other three totally agree with super bullish on agents everywhere. I would actually love to see this. So Thomas Chapman's here says someone will build a system that will let you tune your social timeline algorithm just by expressing your tastes and preferences. That would be great. So like never show me a poster about Blank ever again and you just won't see stuff like that. That would be really cool. I mean, I'd love to see X do that. I feel like a lot of the platforms are disincentivized to do that. Although I believe Threads already allows you to sort of custom tailor your feed a little bit better than the rest of the platforms. Cyber Mage AI says lots of crappy 5 to 10 second AI videos clogging up everyone's timelines. Out of that, there will be some longer ones that are amazing. I mean that's now, that's today, right? You're already seeing a lot of crappy 5 to 10 second AI videos clogging up timelines. But I do agree we will see some longer ones are much more amazing. I've already seen some that are like, you know, two or three minutes that I'm really blown away by. Robotics take off. AI agents 2025 AI agents AI agents 3D AI will continue improving. That's another thing I didn't mention. 3D AI has gotten really good with tools like Meshi and Backflip and Tripo. There's a lot of really, really good AI to 3D object platforms now where you can go and 3D print them or move them into Unreal Engine or pull them into Mixamo and animate them. We've already seen those tools. Tools get pretty good. I do think we're going to see them get really, really good in 2025. And maybe that's the next step that we need to get more AI in the gaming development experience. I don't know. I still think a lot of the gaming world is going to reject it and try to push back hard. I'll go ahead and cut it off right there as far as the Twitter posts, because most of the other posts as I scrolled through them were saying the same kind of stuff. We're going to see AI agents, we're going to see AI video get a lot better. We're going to see leaps in quantum computing. We've got some people that think we're going to see some big leaps in medicine as a result of AI. But one thing I can guarantee is we're going to get a lot more AI slop. And that's one thing that I'm not a fan of when it comes to AI is just a lot of quickly produced junk that people slap up and throw up on social media. And then you get feeds that are clogged with it. We're going to see that in the form of text, video, audio, images, all of it. Not really a fan of that, but we're definitely going to see more of that in 2025. If there's one prediction I can say with 100% certainty, it's that we're going to see more AI slop. But I'm going to do my best over 2025 to keep you up to date with the news that's actually important in the AI world. The advancements that I actually think are useful or helpful or have really big implications to the world. I'm going to continue to make tutorials in 2025 for you. I'm going to continue to break down all the latest tools. We're going to do a lot more fun live streams. I'm going to keep on sharing short form clips with little bits of tips and advice like I've been doing. I'm super excited for what 2025 holds. I'm excited to dig in and get more immersed along with you and just keep looping you in on the best, coolest, most most important news going on in AI and emerging tech. My goal is always to be helpful and show you the areas that I think could be genuinely beneficial to most people or to the world. That's really the goal with the channel. I'm not slowing down in 2025. I'm probably ramping things up and sharing even more cool workflows and tools and tutorials and news and all that kind of stuff. So if you like that kind of stuff, make sure you like this video and you subscribe to this channel and that will ensure that more videos like this show up in your feed. And if you haven't already, check out Future Tools IO. This is where I share all the cool AI tools I come across. I always keep the AI News page up to date on a daily basis. So all the latest AI news is available here. And I've got a 100 free newsletter where I'll share just the coolest tools and most important AI news directly in your inbox when you sign up. You also get free access to the AI Income Database, a little database I made of cool side hustles and ways to make money leveraging AI tools. It's. It's all totally free. You can find it over at FutureTools IO. Thank you once again for tuning in. Thank you so much to Opus Clip for sponsoring this video. I really appreciate you all. Thank you and once again, Happy New Year. 2025 is going to be a fun one. See you in the next video. Bye bye.