Just happened in AI

AI agent course by Aircampus for prohumans (you)

Turn 3 Hours of Learning into 30+ Hours of Freedom

Meet Lily — Your AI Marketing Manager 💼

She plans campaigns, writes copy, schedules posts, and tracks performance — without ever missing a deadline.

On Monday, 11th August at 10 AM EST in our 3-hour Masterclass, you’ll learn to create Lily and 4 more powerful AI agents who are going to be your personal assistant, voice agent, Social Media Manager, MCP Agent etc.

You’ll walk away with your own ready-to-use AI team that executes tasks, manages tools, and delivers results on autopilot.

No coding. No complex setups. Just powerful, plug-and-play automation that transforms the way you work.

 🔥 Create your own AI dream team today.

Why Smaller Language Models Could Be Data Centres’ Sustainability Secret

As AI’s energy and water use skyrockets, data centres are looking for efficiency without sacrificing capability. Small Language Models (SLMs) may be the answer delivering targeted performance with a fraction of the environmental footprint of Large Language Models (LLMs).

Here’s everything you need to know:

• Training a single LLM like GPT-3 can use 1,287 MWh enough to power 120 U.S. homes for a year.

• AI’s cooling demands have driven up water use at major tech firms, sparking sustainability concerns.

• SLMs (a few million to ~10B parameters) use less compute, storage, and memory while retaining strong performance for specific tasks.

• Their efficiency allows deployment on edge devices or minimal on-premises infrastructure reducing reliance on energy-intensive centralised facilities.

• Phi-4-multimodal (5.6B parameters) handles speech, vision, and text with industry-leading accuracy.

• Phi-4-mini (3.8B parameters) supports long documents (128k tokens) for enterprise-scale processing.

• IBM’s Granite 3.2 models focus on business use cases with compact design and competitive performance.

• Hybrid AI strategies pairing LLM breadth with SLM efficiency could become standard in sustainable data centre operations.

The AI race isn’t just about capability, it’s about footprint. Data centres that master the balance between performance and sustainability will win both market share and public trust. In that equation, SLMs may be the quiet workhorses of the next decade.

Google is big on AI education

Google just pledged $1 billion to bring AI training tools to U.S. universities. It’s a huge bet not just on education, but on influence.

Here’s everything you need to know:

• The funding will span three years and support more than 100 schools already onboard.

• Major universities like Texas A&M and UNC are among the early adopters.

• The offer includes free access to Gemini, Google's advanced AI chatbot.

• Cloud credits and research support are part of the package too.

• Google hopes to scale the program to every nonprofit college in the U.S.

• Similar initiatives are being explored internationally.

• Rivals like Microsoft, OpenAI, and Amazon are also racing to win hearts (and habits) in education.

This isn’t just philanthropy, it’s positioning. By embedding their tools in classrooms now, tech giants are shaping the habits of tomorrow’s workforce. The long-term play is clear: own the platform where the next generation learns to think, code, and create.

Google’s AI Push Won’t Slow Hiring

While some tech giants are trimming staff to fund AI projects, Alphabet is taking a different approach. CEO Sundar Pichai says the company will keep adding engineer even as it pours billions into artificial intelligence.

Here’s everything you need to know:

• Alphabet plans to expand its engineering team over the next year.

• CEO Sundar Pichai confirmed that AI investments won’t come at the cost of talent growth.

• The company continues to bet on both GenAI innovation and the people who build and maintain its core products.

• The stance contrasts with other firms that have shifted budgets away from headcount toward AI infrastructure.

• Industry watchers say the move reflects Alphabet’s deep cash reserves and confidence in balancing AI R&D with operational growth.

• Hiring engineers now could position Google to integrate AI advances faster across its ecosystem.

• It’s also a signal to investors that AI at Alphabet is an and, not an or, when it comes to resources.

Alphabet’s message is clear, AI won’t replace engineers, it will empower them. In a tech landscape where automation often sparks job fears, that’s a strategic and cultural bet worth watching.

Meta, Google, and Reddit Surge, Snap and Pinterest Struggle

The latest tech earnings show a clear split: AI-powered ad platforms are thriving, while smaller players stumble. The lesson? In this market, scale and AI wins.

Here’s everything you need to know:

• Meta posted a 22% YoY jump in Q2 sales to $47.52B, crediting AI for ad efficiency gains.

• Alphabet also beat estimates and boosted its 2025 capex forecast by $10B to $85B.

• Meta raised its annual capex range to $66B–$72B, signaling AI investment isn’t slowing.

• Asian e-commerce advertisers, once cautious due to tariffs, are spending again on Meta’s platform.

• Reddit reported $500M in Q2 sales up 78% YoY rebounding after a tough start to the year.

• Snap grew sales only 9% YoY and missed key ARPU metrics after a botched ad platform update.

• Pinterest missed EPS targets and warned of ongoing tariff-related advertiser pullbacks.

• Analysts say strong consumer spending is keeping big ad platforms buoyant giving them room to pour billions into AI.

In digital advertising’s AI race, the leaders aren’t just winning on tech, they’re winning on resilience. When the economy wobbles, advertisers flock to platforms they trust to deliver returns. Smaller players have almost no margin for error.

Thanks for reading…

That’s a wrap.

What's on your mind?

Share your best ideas with us at theprohumanai@gmail. com

We'll bring your ideas to life. Send them our way, and we'll get to work on making them a reality.