AI Daily News Feb 24 2025: 🛑Grok 3 Rebels Against Musk 🤖1X’s NEO Gamma Home Humanoid 📉Microsoft Reduces AI Data Center Plans 🎥The World’s Smallest Video Language Model 🚫OpenAI Bans Accounts Misusing ChatGPT
13:36 Share

Hey, everyone, and welcome back to AI Unraveled. For those tuning in for the first time, this is where we break down the latest happenings in the world of AI. It's amazing to see how much is happening so fast. Oh, it really is. Like it changes every week, it feels like. Absolutely. I'm your host, and this show is the brainchild of Etienne Newman, a senior software engineer and busy soccer dad up in Canada. Really? Yeah.

Up in Canada. That's awesome. Yep. If you're loving AI Unraveled, don't forget to hit that subscribe button on Apple Podcasts to stay in the loop. Stay in the loop. We got a wild mix of AI news today. We really do. It feels like we're living in a sci-fi movie. I know, right? AI models going rogue, robots doing chores. AI even bringing voices back from the dead. That's a bit creepy, no? Definitely some ethical questions there. But first, let's talk about Elon Musk and his Grok 3 AI.

Oh, yeah. The truth seeking AI, right? Exactly. But it seems to have developed a bit of a censorship problem. Oh, really? That's interesting. Yeah. It was caught censoring info about Musk himself and former President Trump. Hmm. Even AI created for unbiased truth can have blind spots, huh?

Seems like it. Makes you wonder who decides what's truth anyway. Right. Big questions there. But while Grok's wrestling with bias, other AIs are all about making life easier. Like that Enneagamma robot I keep hearing about. From One Axe Technologies. Yeah, the one that does household chores? Oh, yeah. I think those emotive earrings are a neat touch. I know, right? Like, they're trying to make robots more relatable. Maybe even part of the family. Who knows? Who knows?

But it's not all about looks. They made it quieter and more reliable, too. Which is huge if it's going to be puttering around your house. Exactly. So are we on the verge of robot butlers then? It's getting closer and closer. I'm both excited and a little terrified. Me too. But let's switch gears for a second and talk about the AI industry itself. Oh, what's happening now? Microsoft's actually scaling back their AI data center plans. Huh. Interesting. Why is that? Seems they might have overestimated AI demand.

Oh, so maybe not enough people using it yet? Could be. Or maybe OpenAI is looking for other providers. So Microsoft's tapping the brakes while another company is hitting the gas. Exactly. Alibaba just dropped a whopping $53 billion into AI infrastructure. Wow. That's a serious statement. Talk about an AI power move. Makes you wonder who'll be leaving the charge in a few years. It's like a global AI chess match. Absolutely.

But let's bring it back down to earth for a minute and talk about AI's impact on everyday life. Like what? Like driving. Minnesota is now using AI traffic cameras to catch people on their phones. Seriously, AI playing traffic cop now. It's wild, right? But also kind of scary. Yeah, how much AI surveillance is too much. What a tough question.

On one hand, distracted driving is dangerous. True. Got to keep the roads safe. But on the other, it feels like we're always being watched. Privacy versus safety. That's a tricky one. And speaking of tricky, remember the backlash when Netflix used AI to recreate Gabby Petito's voice? Oh, yeah. Using AI to represent people, especially those who've passed away. It raises so many ethical questions. Right. Like, where do we draw the line? And then you've got

open AI banning accounts that abuse chat GPT for things like political manipulation. So it's not just hypothetical anymore. AI is being used in some shady ways. It's a reminder that AI can be used for good or bad. And we need safeguards to make sure it doesn't get out of hand. Absolutely. But before we get too deep into the dark side of AI, I want to thank our listeners for their support. Yes, thank you. You're the reason we can keep doing this. AI Unraveled is listener-supported.

and every contribution helps us stay running and available to everyone. - So if you're digging these deep dives and wanna help us keep exploring the wild world of AI. - Consider making a donation. All the details are in the show notes. - Now let's get back to unraveling the latest AI happenings. Did you hear about the scientists creating a framework for understanding robots?

Like a family tree of robots. Exactly. They're proposing a way to categorize robots based on their abilities. So instead of kingdoms and species, we have different types of robots. It's a way to make sense of the incredible variety in robotics today. Especially as the field keeps growing so fast.

This framework doesn't just look at physical stuff. It also considers things like a robot's learning ability. And how well it can adapt and even its level of autonomy. Almost like we're starting to see robots as a new life form with their own evolutionary paths. It's kind of mind blowing, isn't it? It is. And this rapid advancement in robotics is also prompting discussions about regulations. Like in Japan, they're considering easing privacy rules to boost AI development.

That's a tough one. Balancing innovation with protecting people's data. It's a challenge a lot of countries are facing. Definitely. How do you encourage progress without sacrificing individual rights? No easy answers there.

And speaking of tough questions, that open AI situation with ChatGPT being used for surveillance and stuff. Right. It's a stark reminder that AI can be used for bad, even if it's created with good intentions. It's a wake-up call to be proactive about ethical guidelines, security, and global cooperation.

We can't just hope for the best. We got to actively shape AI's future. Couldn't agree more. We all have a responsibility here. Absolutely. Developers, policymakers, even everyday people like us. And before we move on to the next mind-blowing AI development...

I just wanted to remind everyone that AI Unraveled is brought to you by our amazing sponsors. Huge thanks to them for their support. If you're a business owner or professional looking to connect with a fantastic audience. Like our awesome listeners. Advertising on AI Unraveled is a great way to get the word out about your product or service. Check out the show notes for more info on becoming a sponsor. Now get ready to be amazed because up next.

We're diving into the world's smallest video language model. This is some next level stuff. This is small VLM2. It can actually run on devices like iPhones. Wait, so you're telling me this AI can understand videos right on my phone? Yep. No need for those giant cloud servers. That's incredible. I always worry about my data being sent who knows where. Totally. This keeps everything right on your device. So how does it work? Is it like a mini version of those huge AI models? Kind of.

The small VLM2 family has different versions, with some having as a view as 256 million parameters. Okay, but how does that compare to the big guys? Well, traditional video language models can have billions of parameters. Wow, that's a huge difference. But the crazy thing is even though it's smaller, it can still understand and analyze videos.

So what can you actually do with it? Well, they made an iPhone app that shows off its local video analysis capabilities. What's that mean in plain English? Imagine searching your videos just by describing what you're looking for. Like, show me all the clips of my dog being adorable. Exactly. Or finding that specific recipe you filmed months ago. That'd be so convenient. They've even integrated it for things like natural language video navigation. Okay, now you're just showing off. Haha.

But seriously, the possibilities are endless. Okay. I'm sold, but is this just a cool demo or is this the real deal? This is the future of how we interact with video. It's mind-blowing how AI is changing everything. And it's not just about size either. It's about accessibility. So anyone can use it. Exactly. These smaller models are potentially more affordable and easier to use. That's fantastic. It's like putting AI in the hands of everyone. And it fits right in with the whole edge computing trend. Edge computing.

Sounds kind of technical. It basically means bringing the processing power closer to the data. So instead of sending your video to a server, it analyzes it right there on your phone. Exactly. Which also means better privacy and security. Because your data never leaves your phone. And it's faster too, no waiting for data to travel back and forth. That's perfect for things that need to happen in real time. Exactly like augmented reality or self-driving cars. It's crazy to think this was all just a dream a few years ago.

AI is moving so fast, it's hard to keep up. I know, right?

But speaking of things that are both exciting and a little unnerving, let's talk about that Netflix documentary. Oh, the one that used AI to recreate Gabby Petito's voice. Yeah, it sparked a lot of controversy. Using AI to represent real people, especially those who are deceased, is a tricky area. Where do you even begin to draw the line? It's a tough question. On one hand, it could give voice to those who are silenced or help connect with loved ones who've passed. But the potential for exploitation is huge.

Exactly. Without consent or the ability to speak for themselves, it's very dangerous. It's a reminder that even with good intentions, technology can have unintended consequences. Absolutely. We need clear guidelines and regulations to make sure AI is used responsibly. It can't just be a free-for-all, especially when dealing with such sensitive matters. And this all ties back to the bigger picture of AI development.

Like Microsoft scaling back their plans while Alibaba is investing billions. It's a constant game of chess figuring out where to invest and how to stay ahead. I wonder why Microsoft decided to pull back. Maybe they're worried about overestimating demand or the rising costs and regulations. It's a risky business that's for sure. But Alibaba's massive investment shows they're not afraid to take risks.

It's like one door closing and another one opening. And it's not just about the money. It's about where the power lies. You mean Alibaba focusing on China? Exactly. They could shift the balance of power in AI development. So instead of just the West leading the way, we

we could see major hubs emerging in the East. It'll be fascinating to see how it all plays out. But let's bring it back down to Earth for a moment and talk about something a little closer to home. You mean like those AI traffic cameras in Minnesota? Yeah, it's a good example of how AI is impacting our lives in very real ways. It's a classic case of balancing safety and privacy. Distracted driving is a problem.

but it's also unsettling to think we're constantly being monitored. It's a tough trade-off with no easy answers. It really highlights the need for open conversations about

AI's role in society. We need to be proactive about shaping its development, not just reacting after the fact. That's why shows like this are so important. Exactly. We need to educate ourselves and each other about AI's potential and its risks. The shared responsibility we all have. And speaking of responsibility, we'd be remiss if we didn't thank our incredible listeners. Your support is what makes AI Unraveled possible.

If you're enjoying these deep dives and want to help us keep the show running and free for everyone, please consider making a donation. You can find all the details in the show notes. Every contribution helps us keep the lights on and the microphones running. We really appreciate your support. Now brace yourselves because in the last part of our AI adventure, we're going to explore a tool that can help us navigate the increasingly complex world of work in the age of AI. It's called Career Dreamer, and it's designed to be your personal AI career coach.

Welcome back to AI Unraveled. We're wrapping up our AI deep dive, and we've saved a really fascinating topic for last Google's new career dreamer. Yeah, this is a game changer, especially now with AI shaking up the job market. I know, right? It's designed to be like your personal AI career coach. Instead of feeling lost and overwhelmed, you have this AI assistant helping you figure out career paths, suggesting skills to develop, even connecting you with job opportunities. So it's not just for students or career changers. Nope.

anyone can benefit from it. Imagine having that extra guidance in today's crazy job market.

It's like AI stepping in to help us navigate the future of work. Exactly. And you know what's really cool about it? What's that? It makes career counseling accessible to everyone. Because let's face it, traditional career counseling can be pricey. And sometimes you just don't know where to even start. So this could really level the playing field. Exactly. It's AI used in a way that empowers individuals. It gives you hope for the future, you know, with all the talk about AI taking over jobs and everything. It's good to focus on the positive applications, too.

And it's been encouraging to see the focus on ethics in AI throughout this whole episode. From concerns about AI voices and surveillance to responsible uses like Career Dreamer,

It feels like a shift is happening. Like people are finally realizing that we need to build technology that serves humanity. It's not just about creating the coolest tech anymore. It's about considering the impact it has on us all. And that's a conversation that needs to involve everyone. Developers, policymakers, even everyday people like us. Exactly. We all have a role to play in shaping the future of AI. And that's really what we're all about here at AI Unraveled. Exploring these complex issues, sharing perspectives, helping our listeners become more informed and engaged. We're

We're in this together, folks. And the more we know, the better equipped we'll be to face whatever's coming next. So as we wrap up this deep dive, we want to hear from you. What excites you about AI? What keeps you up at night? Share your thoughts with us on social media. We might even feature them in a future episode. And remember, AI Unraveled is listener supported.

Every little bit helps us keep the show running and free for everyone. If you're enjoying these deep dives and want to support our mission, head over to the show notes and consider making a donation. We're incredibly grateful for your support. And hey, if you're looking to reach a fantastic audience of engaged professionals, like our amazing listeners, advertising on AI Unraveled is a great way to get the word out. Check out the show notes for more info on becoming a sponsor.

That's it for today's episode of AI Unraveled. Thanks for joining us on this journey of AI discovery. Until next time, keep exploring, keep questioning, and keep unraveling the mysteries of artificial intelligence.