🧱 Meta Is Building Its Own AI Backbone

+ Build Powerful Simulations with Gemini 3 🧠

In partnership with

Today’s menu 🍽️👇️ 

Length: 5 minutes  

  • Meta Is Building Its Own AI Backbone 🧱

  • Claude Is Becoming Your AI Teammate 🧠

  • Google Adds Buy Buttons to Gemini 🛒

  • Anthropic Launches Claude for Healthcare 🏥

  • Use Gemini 3 to build powerful simulations 💻

  • Prompt of the week 💬

  • Top AI tools of the week ⚒️

Start learning AI in 2025

Everyone talks about AI, but no one has the time to learn it. So, we found the easiest way to learn AI in as little time as possible: The Rundown AI.

It's a free AI newsletter that keeps you up-to-date on the latest AI news, and teaches you how to apply it in just 5 minutes a day.

Plus, complete the quiz after signing up and they’ll recommend the best AI tools, guides, and courses – tailored to your needs.

FRESH AI NEWS

Meta Plans Its Own AI Infrastructure Stack 🧠 

Meta announced that it is creating its own AI infrastructure. This move is intended to lessen its dependence on outside cloud and GPU providers and to support its growing AI goals 🧠

  • Zuckerberg confirmed Meta is investing in proprietary AI infrastructure to support large models and internal AI workloads 🛠️

  • The project involves creating special hardware and improving software, rather than just using large cloud services 💻

  • Meta says the goal is cost-effective, scalable AI compute tuned for its own AI models and services 📈

  • Initial deployment is for Meta’s internal AI teams (LLMs, agents, AR/VR models) before broader rollout 🌍

  • Meta is hiring infrastructure engineers, systems architects, and MLops specialists to build and operate the stack 🤝

Why You Should Care: Meta is building its own AI infrastructure to control performance, costs, and innovation, competing with Google and OpenAI. This reflects a trend where major AI companies seek full ownership of their technology stack, beyond just using rented GPUs.

Anthropic Pushes Claude Into Cowork Mode 💼 

Anthropic launched Claude Cowork, an AI that helps non-developers manage files via conversational instructions 🤝

  • Claude Cowork is designed to feel like you're delegating tasks to a coworker by queuing up work for Claude to execute 👥

  • Users grant Claude access to a specific folder on their Mac, enabling the AI to read, edit, organize, and create files on its own 📂

  • Examples include sorting downloaded files, generating spreadsheets from receipts/screenshots, or producing draft reports from scattered notes 📊

  • Cowork can tie into tools like Asana, Notion, PayPal, and connect to Claude in Chrome for web-related workflows 🔗

  • It uses the same underlying Claude Agent SDK that powers Claude Code, but in a much more approachable, non-technical interface 🧰

  • Anthropic warns about risks like accidental file deletion and prompt injection, stressing that agent safety is still being refined ⚠️

Why You Should Care: This chat feature advances agentic AI from Q&A to task execution, potentially automating office tasks like research, reporting, organization, and app coordination by removing coding barriers.

Gemini Turns Into a Shopping Tool 💳

Google is introducing in-chat shopping and buy buttons in Gemini and AI Mode search, enabling users to discover and make purchases within the AI experience 🛍️

  • Users can now make purchases directly within Gemini chats and AI Mode in Search without being redirected to external sites 🛒

  • Google introduced an open standard (UCP) with partners like Shopify, Walmart, Target, and Wayfair to power seamless shopping from discovery to payment 🌐

  • Over 20 companies, such as Visa, Mastercard, PayPal, Best Buy, and Macy’s, have endorsed UCP to expand checkout options and payment flexibility 🤝

  • Google is also launching business agents that let users chat with retailers in their own voice inside Search/AI Mode 💬

  • Integrations with Walmart, Shopify, and Wayfair mean users can browse and buy from top merchants directly through Gemini 📍

Why You Should Care: Google is adding buy buttons and checkout flows to AI conversations, turning Gemini and AI Mode into full shopping platforms, streamlining the process from inquiry to purchase and challenging traditional e-commerce.

Anthropic Pushes Claude Into Healthcare 🚑

Anthropic launched Claude for Healthcare, a HIPAA-ready AI toolset, soon after OpenAI's ChatGPT Health, underscoring rising competition in AI medical tools🩺

  • Anthropic announced Claude for Healthcare at the JPMorgan Healthcare Conference, adding AI tools for clinical workflows, payer administration, and patient support 🏥

  • Claude can now connect to health systems and major medical databases (CMS, ICD-10, NPI, PubMed) to speed research, coding, and prior authorization tasks 🔗

  • Users can opt in to let Claude access securely held health records or fitness data to interpret labs, trends, and histories 📊

  • The platform is built with enterprise privacy and compliance in mind for hospitals and insurers ⚖️

  • Anthropic leverages its latest Claude model to boost performance on medical reasoning and data-driven tasks 🧠

  • The launch comes just days after OpenAI’s ChatGPT Health, highlighting the competitive push into regulated healthcare AI 📆

Why You Should Care: This marks a major escalation in AI's healthcare ambitions, as Anthropic and OpenAI race to integrate large language models into clinical workflows. Claude for Healthcare emphasizes advanced integrations and compliance, competing with ChatGPT Health in real-world medical systems.

CREATE AI SIMULATIONS

Build Powerful Simulations with Gemini 3 🧠 

How to transform abstract ideas into interactive demos with Gemini 3, allowing you to create simulations, learning tools, and mini-apps in one step for quicker testing and better visual understanding.

Here’s how to get started 👇️ 

  1. Open Gemini and choose the right engine by selecting Gemini 3 Pro for strong logic and explanations, or switching to DeepThink when you want deeper reasoning and more complex outputs (availability may vary by region).

  2. Describe the outcome you want, not the method. For example:

    “Create an interactive visual model that shows how a quantum computer processes information, with explorable components.”

  3. Inspect the generated experience, then use the Explain option to turn the output into a guided walkthrough that teaches concepts while revealing how the system works under the hood. Save or share the generated code if needed.

  4. Try using the prompt for different purposes, like educational games, workflow tools, analytics dashboards, or visual planners. If the first version isn't clear or easy to understand, make changes and try again.

Pro Tip: Ask Gemini to solve the same problem in two different ways (visual vs. functional). Comparing outputs sharpens intuition and exposes better design and reasoning patterns you can reuse elsewhere.

The best place on the Internet to learn about AI! 🌐 🤖 

I just started my own X page where I provide AI news, guides, hacks and more (for FREE)!

We don’t miss a day, that means you get the most recent AI news, practical workflows, expert guides and premium content everyday.

TOOLS OF THE WEEK

These AI tools are going VIRAL 🐝 

 

GLM-4.7 - Z.ai's new SOTA open-source model 🤖 

SAM Audio - Meta’s model to separate sounds using text prompts 🚀 

GPT 5.2 Pro - OpenAI’s most advanced frontier model 🧮 

Grok Imagine - xAI's image and video generation platform 🌌

Warm regards,
Leo Grundström / Founder of Daily AI Edge

Reply

or to participate.