Hey, have you ever used an AI tool and afterwards felt a little fuzzy, like your own brain kind of took a backseat for a bit? Oh, definitely. It's a weird sensation, right? It really is. Well, today we're going to take a deep dive into that very feeling. We're exploring this groundbreaking MIT study that really unpacks what AI might be doing to our thinking. But we're also balancing that. Exactly. Balancing that with some truly mind -blowing advancements happening in AI itself.
We've pulled together a stack of sources for you from that key MID research to all sorts of updates and breakthroughs. It's going to be a lot to explore. Yeah. What's so fascinating here is just the sheer speed of change in AI right now and what that ultimately means for you, the
listener. We're talking about, sure, profound opportunities, absolutely, but also some really nuanced challenges, especially to our... cognition right this deep dive isn't just about what AI can do though believe me we'll get into some incredible new tools it's also about what it does to us and how we adapt or how we should adapt maybe okay let's dig into this let's call it concerning finding from MIT It's all about what chat GPT or, you know, LLMs in general might
be doing to our brains when we use them for stuff like writing essays. And the core finding, it's kind of counterintuitive, like you said. Yeah. AI actually made people better at the task, at writing, but simultaneously maybe worse at thinking. Sounds a little wild, doesn't it? It does. So how did they even figure this out? What was the setup? Well, the study was pretty rigorously designed, actually, to isolate these effects. MIT brought in 54 participants over four months,
all focused on essay writing. Okay. They split them into three groups. One group used ChatGPT. Another used Google, kind of like, you know, how we might have done research back in the day, just basic search. Right, 90s kid style. Exactly. And then the third group, totally analog. Just their brain, pen and paper, no digital help. Interesting. And the results? They were really striking. The ChatGPT group showed a significant drop in brain connectivity, measured by EEG,
you know, mapping brain activity. I mean, it literally flatlined in some areas during the writing test. Flatlined? Seriously? Yeah. And what was even more telling was the follow -up. When these same ChatGPT users were asked to write without AI afterwards, They performed like novices, like people who hadn't been practicing writing for months, which really highlights the dependency, you know? Wow. So the brain's literally like disengaging. That's pretty intense. It really
is. And get this. Here's a truly astonishing statistic from the study. Okay. 83%. 83 % of the chat GPT users couldn't quote a single line they had just written using the AI. 83%. Compared to what? Compared to only 11 % of the brain -only group. Oh, man. It's almost like if you let AI do the heavy lifting, your brain just goes, cool, I'm clocking out now. That's kind of terrifying, actually, because here's where it gets really interesting, you know, the maybe creepiest part.
The users themselves felt fine. They didn't report any mental fog, no perceived decline, nothing. Right. The subjective experience was okay. Exactly. The only way this drop was visible was on those EEGs. Yeah. It's like this invisible trade -off, isn't it? You get efficiency, maybe better output sometimes, but at a hidden cognitive cost. Yeah. And MIT even coined a term for this, cognitive debt. Right. Think of it like the Google Maps effect. Yeah. The more you rely on GPS. The worse
your own sense of direction gets. Exactly. You're essentially borrowing against your future cognitive capacity for that short -term convenience. That's a heavy concept. It is. But, and this is crucial, the study is absolutely not anti -AI. That's a key takeaway. Okay, good. Because they actually discovered a brain -friendly AI workflow within the study. So it raises this important question for all of us. How do we engage productively? How do we use these tools without racking up
that debt? But what did they find? It was a brain -friendly way. The optimal workflow they identified was first think, really process the ideas yourself, then write your own draft, and then only then use chat GPT to maybe polish, refine, or get feedback. Ah, so the AI comes in later. Exactly. And what's truly fascinating here is that people who used this method, they actually showed stronger brain connectivity than anyone else in the study. Stronger than even the brain -only group. Stronger
than the brain -only group. That's a huge insight, right? It suggests thoughtful AI integration can actually enhance cognition. Wow. But it really highlights that if you just rely on AI from the start, letting it do the initial heavy lifting, well, your cognitive core can, you know... get a bit mushy. Right. So if you're still in that phase of, oh, AI can just write it for me, maybe rethink that. Let your brain take the wheel. At least sometimes your hippocampus will thank
you later. That's such an important distinction. It's not about avoiding the tool. It's about how you use it thoughtfully. Precisely. But hey, it's not all about potential brain drain, right? There's some incredible stuff happening out there, too. Let's kind of zoom out a bit. Talk about the broader AI landscape. What's been buzzing? Yeah, there's a lot going on. For sure. Take AI video models. Just recently, we saw ByteDance's Seedance 1 .0 and then this kind of sneaky new
contender, Hailu O2. Oh, yeah. They jumped up the leaderboards fast. Right. Leapfrogging models like Google's VO3 almost overnight just shows how dynamic this space is. Innovation cycles are shrinking like crazy. Wild. And speaking of VO3, apparently clever TikTok users. are racking up millions of followers creating ASMR videos using its new native audio features. Interesting use case. Yeah. VO3 seems to be like a goldmine
for creators right now. Oh, and get this. You can even generate images with ChatGPT just by texting a specific 1 -800 number on WhatsApp. On WhatsApp. Really? Yeah. Don't even need an account, apparently. You get one free image daily. Have you tried that yourself? I haven't tried the WhatsApp thing yet, no. But I've seen some impressive outputs people are sharing. It just speaks to how accessible these tools are becoming,
you know? Totally. But if we connect this all back to the bigger picture, it's not just exciting new toys. There are also some really significant considerations, things we need to be aware of. Like what? Well, for instance, Jeffrey Hinton, you know, often called the godfather of AI. Yeah. Heard of him. He's openly stated there's maybe a 10 % to 20 % chance that AI will displace humans completely in some sectors. He even mentioned specific fields he thinks face a heavier risk,
saying they should be terrified. Wow. That's sobering. It is. It's not necessarily a prediction, more like a stark warning about the potential scale of disruption. And then you has Reid Hoffman, LinkedIn's co -founder. What's his take? What's fascinating about his advice is he's saying things like prompt hacks and what he calls vibe coding, you know, just guessing prompts without real understanding. He says that's not going to be
enough to really thrive. He argues it's going to come down to actual judgment, critical thinking, applying these tools strategically. That's what will differentiate people. That makes sense beyond just tinkering. Right. And on a really practical note, something you definitely need to keep in mind is data privacy. Ah, yes. Important. Our sources indicate that a pretty significant chunk, like 63 % of ChatGPT user data, actually contains personally identifiable information, PII. 63%,
that's high. It is. Yet only about 22 % of users are even aware they can opt out of data collection. And crucially, your inputs and outputs, they're being saved indefinitely. Indefinitely. Okay, that's something everyone should know. Definitely something to be aware of, you know, as you interact with these platforms. Absolutely. That privacy point is huge, knowing what's happening with your data. Yeah. But, okay. On a more positive note for specific industries, we are seeing some
amazing efficiency gains, right? Oh, for sure. Like Nobla AI. This startup just got $70 million in Series C funding. Right, for the clinical documentation tools. Exactly. Used by like 85 ,000 clinicians, and it cuts their documentation time by over 50%. That's pretty huge for healthcare efficiency, isn't it? Massive. Reducing burnout, freeing up time for patients. Big potential there. And then you look at the big players. Amazon is pouring a massive $100 billion into automation.
Yeah. Over 1 ,000 AI projects underway. $100 billion. Wow. Yeah. But interestingly, they've also announced some AI hiring freezes in certain areas. It just shows how rapidly this stuff is transforming operations across the board. Big shifts happening. Okay. But here's where it gets really interesting again. Let's talk about a major breakthrough. Chinese AI startup, Minimax. Ah, yes. Minimax and M1. They've kind of changed the game with their M1 model. We're talking major
leap forward in AI capabilities here. Definitely pushing boundaries. It's the most impressive feature. Get ready for this. A one million token contest window. One million. That's huge. It's the biggest we've ever seen. Like, wow. Just massive. The amount of information it can hold in its working memory. At once. Game -changing potential. And here's the kicker. Get this. This whole model. Trained in just three weeks. Three weeks. That's incredibly fast. For under $540
,000. Under half a million? Yeah. Like not even close to a million dollars. Yeah. It's pretty incredible, right? Makes you really wonder about development costs and accessibility going forward. That efficiency is almost as impressive as the context window itself. So connecting this to the bigger picture, the significance of a 1 million token context window is hard to overstate. Explain
that. What does it really mean? It means instead of having that short term memory loss you see in a lot of current models, you know, where they forget the start of the conversation. Right. Very frustrating. This thing can hold an entire project or a whole book or like a massive code base in its head and still reason through it coherently. It basically gives the AI a truly expansive working memory. OK, so much more capable
for complex tasks. Exactly. Our sources show it performs exceptionally well across the board, but it really shines in areas like software engineering. using tools as an agent and obviously that long context reasoning. And how did they do it so fast and cheap? Well, what's fascinating here is how they achieved it. Minimax used a new reinforcement learning method they developed called CISPO. Yeah. Essentially, it's just a smarter, more efficient way for the AI to learn from trial
and error. This led to a 2x faster training speed without needing, you know, those colossal GPU farms or mega budgets. So cleverness over brute force. Kind of proved that clever architecture and optimization can actually beat sheer compute power sometimes, you know. Which is encouraging. Definitely. And the implications. Well, this kind of long context capability is what serious
AI agents really need to become practical. Think about agents that don't forget what you said five prompts ago or agents that can scan hundreds of pages of legal documents or medical records before accurately answering your questions. Right. Handling human scale data. Exactly. It truly opens the door to like personalized memory systems, much smarter assistants, reliable legal or medical AI. This is a. pretty transformative for practical
applications down the line. So, wow. We've really gone on a journey here from the subtle kind of unseen ways AI might be like changing our brains. The cognitive dead idea. Yeah. To what's buzzing in AI tools right now, the good and the concerning, all the way to this massive breakthrough in long context models with mini Macs. It's a lot. It is. It's clear AI is advancing incredibly rapidly on so many fronts and understanding its impact,
both the good and the. potentially challenging is just crucial for all of us navigating this absolutely ultimately I think this deep dive really shows us AI's immense potential but also that vital need for intentional, thoughtful engagement. It's about how you choose to integrate it into your workflow, into your life, not just letting it passively take over or accepting its influence without question. Right. Being deliberate. Yeah. Which raises an important question maybe for
you listening. How does this information shape your approach? How will you think about learning and working with AI going forward? Something to definitely consider. Exactly. It's not about being anti -AI, not at all. It's about being smart about how we use these incredibly powerful... Powerful tools. Well said. So maybe the next time you reach for an AI tool, just pause for a second. Ask yourself, is this enhancing my thinking or is my brain kind of clocking out
here? Think about it. We'll catch you on the next deep dive.