Welcome, AI Pioneers!

Google’s agentic Gemini now books appointments for you. Meta is wiring gigawatts into Prometheus. Nvidia quietly walks back its China retreat. And a North Carolina lab just made chemistry 10X faster. The future isn’t waiting. It’s dialing in.

📌 In today’s Generative AI Newsletter:

  • Google’s Gemini AI now calls businesses for you

  • Nvidia reboots China chip sales amid trade tension

  • Meta’s Prometheus data cluster will rival Manhattan

  • Self-driving labs make 10X leap in materials science

Special highlight from our network

AI Trust Hub: Get Future-Ready. Master AI Risks.

Built by the team behind enterprise-grade GenAI security deployments, 

AI Trust Hub is where real-world experience meets practical upskilling.

Whether you’re a CIO, security architect, or aspiring AI risk leader, our tailored online courses help you confidently tackle the evolving risks of GenAI.

Our self-paced modules blend hard-won field insights with actionable guidance - no fluff, just what works in production environments.

  • Gain deep, practical knowledge of AI risk and policy frameworks

  • Learn to navigate and implement ISO 42001, EU AI Act, and NIST 600

  • Explore topics like AI vulnerabilities, risk scoring, and governance

  • Upskill for career growth in an AI-driven world

  • Designed for both executives and hands-on practitioners

🎯 Built by practitioners. Trusted by AI security teams.

🚀 Launch Offer: Use discount code TRUST50 for 50% off (limited time)

📞Gemini Expands: Google's AI Now Calls for You & Deep Dives

Image Credits: Google

Google is rolling out its agentic AI-powered business-calling tool to all U.S. users, letting Search place calls on your behalf to collect info from local businesses. At the same time, it’s upgrading AI Mode with Gemini 2.5 Pro for subscribers and debuting a Deep Search tool designed for complex research tasks.

Here’s what’s new:

  • AI call agent rollout: Now live for all U.S. users, the feature calls businesses to ask about services, pricing, and availability

  • User-driven prompts: After searches like “dentists near me,” AI asks follow-ups and returns answers without human interaction

  • Gemini 2.5 Pro added: AI Mode now lets Pro and Ultra subscribers select the 2.5 Pro model for better math and logic tasks

  • Deep Search launched: A new research tool that compiles info from hundreds of queries into a cited, structured summary

Google’s moves deepen the integration of agentic AI into everyday services from customer support to search. The bigger question now is how other platforms like Perplexity and ChatGPT will respond, and how consumers will adjust to a more hands-off digital experience.

🌏 Nvidia Reopens China AI Chip Sales Amid Rare-Earth Trade

Image Credit: Adek Berry | Afp | Getty Images

Nvidia is walking back its China retreat. Just weeks after signaling an exit, the company is now applying to resume sales of its H20 AI chips which are directly linked, officials say, to ongoing trade talks over rare-earth elements that China dominates.

Here’s what’s behind the sudden reversal:

  • The trigger: rare-earth leverage: Nvidia’s H20 sales are part of broader negotiations over Chinese control of rare-earth elements used in EVs, chips, and batteries.

  • A fourth-best concession: Lutnick downplayed the chip’s capabilities, telling CNBC China would only receive Nvidia’s “fourth best” hardware

  • Pushback from lawmakers: Critics say the move undermines the administration’s tech export strategy. Congressman warned it risks aiding foreign adversaries

  • Policy vacuum: With the Biden-era AI Diffusion Rule revoked in May and no new framework yet in place, the U.S. stance on chip exports remains murky

This moment reveals the uneasy tradeoff between national security and industrial supply chains. With rare-earths as leverage, China is forcing a new kind of AI diplomacy and Nvidia’s U-turn won’t be the last.

🏗️ Zuckerberg Plans Manhattan-Sized AI Data Center for Meta

Photograph: Jeff Chiu/AP

Meta is going all in on AI, with CEO Mark Zuckerberg announcing plans to build a multi-gigawatt data center so large it will rival the footprint of Manhattan. Dubbed Prometheus, the first of Meta’s next-generation AI superclusters is set to go live in 2026, with another, Hyperion, to follow.

Here’s what’s scaling up:

  • Prometheus will be Meta’s first gigawatt-plus data center, with Hyperion designed to scale to 5 gigawatts

  • Dozens of titan clusters are under construction, each covering a chunk of real estate comparable to entire city blocks

  • Zuckerberg claims Meta will be the first to bring an AI supercluster online, citing a report from SemiAnalysis

  • Superintelligence Labs will lead the charge, run by Alexandr Wang and Nat Friedman after Meta’s $14.3B investment in Scale AI

With nearly $165B in annual revenue, Meta is flexing its ad business muscle to fund a moonshot vision for AI. But with setbacks in Llama 4 and a brain drain to rivals, the company is betting its empire on scale measured in gigawatts and billions.

⚗️ Self-Driving Labs Just Hit 10X Speed in Materials Discovery

Image credit: Milad Abolhasani.

A new system from researchers at NC State makes self-driving labs 10 times more efficient, pushing materials discovery into overdrive. By ditching stop-and-go experiments for nonstop, dynamic chemical flows, these AI-powered labs are now faster, leaner, and more sustainable.

Here’s what makes it a game-changer:

  • Continuous experiments: New dynamic flow system eliminates downtime by analyzing reactions in real time, capturing data every 0.5 seconds.

  • 10X more data: The always-on system streams rich, continuous reaction data, generating 20 data points where previous setups got one.

  • Smarter, faster learning: Real-time data supercharges the lab’s machine learning brain, enabling optimal material discovery in fewer trials.

  • Lower waste, lower cost: With less chemical use and fewer failed experiments, the system shrinks environmental and financial impact.

This shift marks a new phase in AI-led science. As lab automation accelerates, breakthroughs in clean energy, biotech, and materials could arrive in weeks, not years. The tools of discovery are changing and they’re training themselves.

🚀 Boost your business with us. Advertise where 12M+ AI leaders engage

🌟 Sign up for the first AI Hub in the world.

Reply

or to participate

Keep Reading

No posts found