Hi folks,

here are the 10 AI tech and business news stories that I found most interesting since the start of the year.

#1 Meta’s Massive Infrastructure Bet

Meta is scaling its data center capacity to reach « Gigawatt-scale » and achieve Personal Superintelligence.

zuck on threads
  • Infrastructure Surge: Plans to deploy dozens of gigawatts this decade, aiming for hundreds of gigawatts in the long term.
  • Massive Investment: Committed $600B in U.S. infrastructure spending through 2028.
  • Nuclear Power: Locked in 20-year nuclear energy agreements to ensure sustainable and stable power for its AI roadmap.

👉  More details here

 

#2 Microsoft’s « Community-First » AI Infrastructure

A new approach to data center deployment that prioritizes local community needs and sustainability.

  • Social Impact: Aims to minimize environmental footprints (water/energy) while maximizing local benefits.
  • 5 Key Commitments: Affordable/responsible electricity, sustainable water management, local job creation, local tax contributions, and AI education funding.

👉  More details here

 

#3 Anthropic Launches « Claude Cowork »

Claude evolves from a chatbot into an agentic « digital collaborator » capable of executing complex tasks.

  • Autonomous Mode: An agentic feature for Claude Desktop (macOS) available for Claude Max subscribers.
  • Rapid Development: Built in just 1.5 weeks using Anthropic’s own agentic coding tools.
  • Key Capabilities: Direct file access (read/write), multi-step execution without intervention, and automated document/report generation.

👉  More details here

 

#4 Slackbot: The Context-Aware Assistant

Slack transforms its integrated bot into a high-context AI agent for the enterprise.

  • Workplace Integration: Operates directly within Slack channels and threads.
  • Deep Context: Leverages internal conversations, files, and project history to provide personalized support.

👉  More details here 

 

#5 Nvidia & Groq: A $20 Billion Strategic Move

Nvidia secures a massive deal with chip startup Groq to strengthen its hardware dominance.

  • Tech Licensing: Includes a technology license and a significant transfer of specialized engineering teams.
  • Competitive Edge: Highlights the intense race for AI inference hardware and specialized chip architectures.

👉  More details here

 

#6 Google Introduces Universal Commerce Protocol (UCP)

A new open-source standard designed to enable seamless « Agentic Commerce. »

  • Unified Language: Creates a common protocol between AI agents, e-commerce platforms, and payment providers (Visa, PayPal, Google Pay).
  • Seamless Shopping: Allows AI agents (like Gemini) to search, compare, and complete purchases automatically.

👉  More details here

 

#7 The 7-Month Doubling Rule

Global AI compute capacity is growing at an unprecedented rate, doubling every 7 months.

exp growth
  • Exponential Growth: Total available compute has increased 3.3x per year since 2022.
  • Market Share: Nvidia dominates with over 60% of global capacity, followed by custom chips from Google (TPU) and Amazon (Trainium).

👉  More details here

 

#8 Amazon Launches Alexa+ for the Web

Amazon’s AI-powered assistant moves beyond smart speakers to directly challenge ChatGPT and Claude.

  • Web-First Interface: Alexa.com now offers a browser-based chatbot for research and planning.
  • Third-Party Agents: Partnered with Expedia, Yelp, and Uber for autonomous booking and services.
  • Usage Surge: Engagement reportedly increased 3-5x since the AI-infused « Alexa+ » rollout.

👉  More details here

 

#9 Nvidia’s « Alpamayo »: The ChatGPT Moment for Physical AI

Announced at CES 2026, Alpamayo brings real-time reasoning to autonomous vehicles.

  • VLA Model: A 10B parameter Vision-Language-Action model that « reasons » through complex driving scenarios.
  • AlpaSim: An open simulation framework to refine driving logic in realistic virtual environments.
  • Open Data: Release of massive physical AI datasets to accelerate industry-wide development.

👉 More details here

 

# 10 Pro Tips: Engineering at Scale with Claude Code

Anthropic’s Boris Cherny reveals how the pros use agentic coding tools.

boris cherny
  • Parallelization: Running up to 15 Claude sessions simultaneously for different sub-tasks.
  • Institutional Memory: Using a CLAUDE.md file to turn past mistakes into permanent instructions.
  • Sub-agents: Treating sub-agents as « reusable workflow atoms » to automate repetitive dev cycles.

👉 More details here 

That’s it for this time, have a good week-end!