Relax… GPT-5 Isn’t Taking Over the World


Hi Reader,

OpenAI dropped GPT-5 a couple of days ago, and the internet lit up. Some are calling it brilliant. Others are calling it a major let down after all the hype.

After sifting through dozens of tweets and blog posts, here’s my takeaway: GPT-5 isn’t AGI - not even close. It’s more of an incremental upgrade than a godlike breakthrough over previous models. It still makes silly mistakes, has a glitchy “routing mechanism,” and struggles with understanding and generating images.

Case in point: I gave GPT-5 a simple prompt: “Give me 5 watercolor sketchbook ideas with visual examples.”

In one try, it gave me a bunch of text with no visual images, despite saying "Here are some ideas for your sketchbook, each paired with a visual example to spark your creativity". In other tries, it produced just one single image with no explanation 🤷‍♀️. I guess this inconsistency is related to the routing issue mentioned above.

Gary Marcus, one of the most outspoken skeptics of LLMs (and OpenAI's promises in general), declared that “scaling is officially broken.” His blog is worth a read, especially if you’re worried AGI is going to take over the world in 2 years 🙃.

His main argument is that LLMs can’t truly generalize beyond what they’ve been trained on - no matter how large the model is, and their “reasoning” is just a shallow imitation of real thinking.

I first came across Marcus while reading Empire of AI by Karen Hao. The book opened my eyes to the behind-the-scenes world of AI’s power players. I broke down the key ideas in my latest Youtube video.

If you’re curious, you can watch it here:

video preview

So, in short: AGI isn’t here yet! We're safe. 🙂

Wishing you a wonderful week,

Thu 🤗

P.S.: Work with me:

If you've wanted to learn how to code in Python, build AI applications from the ground up and understand all what’s going on under the hood - I’ve built a program to teach you all the fundamentals of Python, Machine learning and AI👇.

You'll join a community of 300+ learners who are building their projects while getting direct access to me and supporting each other along the way.

🔗 Learn More


Thu Vu

Say hi 🙌 on Youtube, LinkedIn, or Medium

Thu Vu

Join 6,000+ data professionals who are advancing their careers by learning from real-world projects, gaining first-hand experience, and accessing top resources delivered to your inbox every week.

Read more from Thu Vu

Hello Reader, In 2016, I graduated with my master's degree. At the time, I only knew a bit of coding in R from working on my thesis. Yet even that modest skill was my major selling point I'd make sure to mention in every job interview! Fast forward to 2026: LLMs can now generate working Python code in seconds. It makes that situation from 10 years ago feel almost ancient. Honestly, it's a little embarassing how I felt like GOD knowing just some basic programming back then 🤣. AI is reshaping...

Hello Reader, How's 2026 started for you so far? Over the past year, I’ve seen a lot of people fall into the vibe-coding trap. I get it. You skip the fundamentals and jump straight into building things with AI. The problem is, AI tools generate 80% of a project really fast. It feels amazing. But then something breaks. Progress slows. You hit a wall. And that last 20% turns out to require way more understanding than you expected. You end up with impressive-looking code you can't fully explain,...

Hi Reader, Just popping into your inbox to wish you and your loved ones a happy, healthy 2026 💛 You know that feeling where you're convinced you did absolutely nothing all year... then you actually look back and go "wait, I did a bunch of stuff"? I kept telling myself I wasn't accomplishing anything for most of 2025. But when I finally sat down and wrote it out, here's what my year actually looked like: Travelled to Japan Made 14 YouTube videos (including 3 data science + AI tutorials and 2...