TOGETHER WITH MAMMOUTH.

Hello again, human brain, and welcome back to your daily munch of AI news.

Here’s what’s on the menu today:

• Sorry AI, humans still matter🤦‍♂️

  CEO is hiring after replacing his team with AI.

• Your city now trains robot armies🧠

  Jim Fan shared his vision for AI agents.

New here? Subscribe! 😎

Peep today's Spot the AI at the bottom. 👇

MAIN COURSE

Humans aren’t obsolete. 🤦‍♂️

Meet Wes Winder. The software developer who decided his human coders were “obsolete” and fired them all. He said he could “ship 100X faster with code that’s 10X cleaner” with AI.

He even blasted this flex:

Bold, right? Except now, he’s crawling back on LinkedIn, publicly hunting for web developers. Redditors are having a field day. One user snarked, “So he’s basically an AI middleman. Can’t anyone else just tell AI to do everything?”

What went wrong?

While AI can churn out code samples or solve routine bugs, it flops when it comes to building massive, coherent systems that real dev teams handle every day. GPT-4 by itself can’t replicate the creativity, problem-solving, and strategic thinking that humans bring to the table.

Now, Winder’s a poster child for the tech world’s over-reliance on AI. In the rush to automate, many forget that human engineers are still the backbone of software innovation. One frustrated commenter nailed it: “AI makes engineers more productive—it doesn’t replace them.”

But hey, who needs devs, right?

Apparently, Winder does—badly enough that he’s willing to eat crow on LinkedIn. This is a cautionary tale for every startup founder who thinks they can ditch their entire team for some fancy chatbot that writes “clean” code. Ultimately, even the best AI is only as good as the humans guiding it.

SPONSORED BY MAMMOUTH.

Get the best AIs in one place for $10.

Mammouth gives you access to the best LLMs: GPT-o1, Claude, Gemini 2, GPT-4o, Grok, Mistral, Llama.

And the top AI image generators: Midjourney, Flux, DALL-E, Stable Diffusion.

For only $10.

SIDE SALAD

AI-powered hive mind?🧠

On Wednesday, NVIDIA's Jim Fan shared his vision for AI agents. He believes most robots will learn in virtual worlds before working in reality.

What's happened?

Tokyo released a detailed 3D model of the entire city for free download, showing how real spaces are going digital.

How does it work?

Virtual cities and spaces train an “iron fleet” of simulated robots that share collective intelligence.

How will these robots learn?

1. Train in massive virtual environments.

2. Share knowledge through a hive mind.

3. Work together on physical tasks.

4. Move to real world once ready.

What's NVIDIA doing?

They designed their Santa Clara HQ in Omniverse (their graphics platform) before building it physically.

Why virtual first?

It's safer, faster, and cheaper to train robots in simulation before deploying them in reality.

YOUR DAILY MUNCH

Recommended Reading 📖

There’s a reason 400,000 professionals read this daily.

Join The AI Report, trusted by 400,000+ professionals at Google, Microsoft, and OpenAI. Get daily insights, tools, and strategies to master practical AI skills that drive results.

Sign up now for free and work smarter, not harder.

Think Pieces 🧠

Google Product lead gives AGI predictions. He claims it will appear as a product release rather than a sudden historic event, with competitors emerging almost simultaneously.

Startup News 💰

Google held a 2025 executive strategy meeting. The CEO told employees that 2025 would be a critical year focused on scaling Google’s Gemini AI assistant for consumers.

Research 👨‍🔬 

3DGraphLLM — uses a 3D scene graph (including semantic object relationships) to create a learnable representation for a large language model.

Orient Anything — learns robust object orientation from 3D models and renders them for label generation. It achieves strong zero-shot orientation estimates from single images.

MEMES FOR DESSERT

SPOT THE AI

3 of these are real coffee mugs. 1 is fake. ☕️

Which one is AI-generated? 👇

Ideas? Comments? Complaints?

Respond to this email or hit me up on 𝕏.

Until next time 🤖😋🧠