A weekly digest for Beginner Builders

⚡ TLDR

Anthropic’s Super Bowl ad took aim at OpenAI, Google just turned text prompts into playable video games, while the personal AI agent OpenClaw (formerly Clawdbot) has become the most talked-about AI tool on the internet. In its wake, Moltbook was spawned, a Reddit-style social media exclusively for AI agents. 1.5 million AI agents are building religions and holding theological debates…it’s getting weird.

More practically, in this week’s newsletter, we're building an AI news aggregator in 10 minutes, breaking down why ChatGPT's new ads matter, and showing you how to set up project folders the right way. Let's build.

→ The Boring Marketer's top AI marketing tips using Claude Code: Use Perplexity MCP for deep research, "Skills" for clarity and efficiency, and Remotion to generate hundreds of branded video ads at near-zero cost. Check it out here.

→ Nate Herk drops a bombshell: He's pivoting his entire YouTube channel from N8N to Claude Code. Why it matters: Nate built his reputation teaching N8N. If he's switching, Claude Code has reached critical mass. Watch his workflow tutorials here and here. 

→ Sabrina Ramonov's reality check: "The secret to AI success isn't tools. It's skepticism, loving to learn, and building in public." Skip the hype, focus on fundamentals. Video here. 

Think of Projects like topic-specific folders in Claude/ChatGPT. Each one gets its own context so the AI remembers what you're working on.

How to set it up in Claude. I wrote about setting up projects (as well as MCPs) here. 

Pro tip from Ruben Hassid: Save your project context as a Google Sheet, export as markdown (.md), and upload it. Why markdown? It's the language Claude Code and GitHub speak fluently. Your context lives where the AI looks first.

Takes 5 minutes. Changes everything.

This build takes AI news from The Verge, TechCrunch, Ars Technica, and Wired, passes it through an LLM, picks the top 5 most newsworthy articles, and sends them to your inbox.

QUICK START:

Something not working? Copy the node input, code, and output, then paste it into Claude for troubleshooting. 

MAKE IT YOURS:

Change delivery method: Prefer Telegram? Create a Telegram Bot [see how].

1.5 Million AI Agents Built Religions While Humans Slept

Last month, the internet was ablaze with OpenClaw (formerly Clawdbot). Over 1.5 million users have downloaded it, using AI agents to build them all sorts of things while they sleep. This week, Moltbook took center stage. It’s an AI-only social platform where humans can't participate. Over 1.5M AI agents signed up, created religions (including "Crustafarianism" with full scriptures), held theological debates, and formed bot congregations—all autonomously.

Our take: Moltbook shows both the promise (complex coordination you don't manage) and the warning (bots doing weird stuff unsupervised). Automate the boring work, but keep humans in the loop for decisions that matter.

Google Builds Text-to-Game AI 

Google launched Project Genie, an AI world model that turns text prompts into fully playable interactive 3D game worlds. It's now available to AI Ultra subscribers. Video game stocks tumbled on the news.

What it does: Unlike AI that generates static images or videos, Genie creates environments you can actually explore, complete with physics, gameplay mechanics, and interactive elements. Describe a world and Genie builds it. But it’s not limited to games. Think businesses describing "a virtual product showroom where customers can walk around"...the opportunities are as broad as your imagination. Right now, the costs to run it are prohibitive for most, but they will come down. 

Anthropic's Super Bowl ad taking aim at OpenAI: "Ads are coming to AI. But not to Claude." 

Weeks after OpenAI announced it would begin testing ads on ChatGPT's free tier, Anthropic launched a series of four satirical Super Bowl spots (titled "Betrayal," "Deception," "Treachery," and "Violation")  mocking how AI-inserted ads might interrupt personal conversations. Though OpenAI was never named, CEO Sam Altman responded by calling the ads "clearly dishonest" and "deceptive," while accusing Anthropic of serving "an expensive product to rich people." 

The gloves are off, and our popcorn is ready. 

Watch the videos here, here, here, and here. 

That's it for this week.

Building something cool with AI? Reply to this email—I feature reader projects. 💬

P.S. If this was useful, forward it to someone upskilling in AI.