97% of Gen Z is using AI to cheat

👋 Good morning/evening (wherever you are). It’s Tuesday.

97% of Gen Z students say they’ve used AI to cheat on essays, homework, test prep, even college applications.

That stat comes from a ScholarshipOwl survey of 12,811 students in May 2025.

It’s a striking number.

Not because students are cheating, but because it shows how deeply AI has already embedded itself into a generation’s daily workflow.

Now zoom out.

Globally, less than 5% of people use AI tools every day.
(And as I write this, that number is climbing toward 10%, but still far from mass adoption.)

There are two milestones that will change everything:

1. When daily AI use hits Instagram levels (~27% of the world)

2. Then comes the smartphone threshold (~60% of the world)

If AI reaches those levels, it moves from social trend to behavioral shift to global infrastructure.

If you think about it, it’s already happening.

Apple started with iOS 18, bringing ChatGPT into Siri and core apps.

With iOS 26, AI is showing up in screenshots, messages, translations, and more.

On the Android side, Google is embedding Gemini deeper into Pixel devices and Samsung is rolling out AI features across its Galaxy lineup.

So yes, technically, most people are already using AI.

But meaningfully? Not yet.

We’re still early. Some will wish they paid attention when it still felt optional.

Remember this:

AI won’t just be something we try. It’ll be something we live in.

OK let’s keep going ↓

🔥 Creative pulse

This is the greatest AI-generated opening sequence I’ve ever seen.

It’s from Kavan Cardoza (Phantom X). He’s one of the few filmmakers I regularly share work from.

Watch the video.

💡 Here’s what you should know

• OpenAI commits $10M to train 400,000 teachers in AI

  OpenAI and the American Federation of Teachers launch a national academy to help K-12 educators use and shape AI in the classroom, with support from Microsoft and Anthropic.

• Moonvalley’s AI video tool for filmmakers is now public

  Trained on licensed footage and built for precise creative control, Marey is one of the first production-grade AI video models now available by subscription.

• China plans 115K Nvidia AI chips for desert data centers

  China is building massive data hubs in Xinjiang and reportedly looking to use 115,000 Nvidia chips in a major push to scale its AI capabilities.

• IBM unveils Power11 server for AI workloads

  Delivers zero planned downtime, fast ransomware detection, and built-in AI acceleration.

• Replit teams up with Microsoft for enterprise app building

  Replit is now in Azure Marketplace, letting business users build and deploy apps using natural language without coding experience.

• Shoppers turn to AI for Prime Day deals

  Adobe predicts traffic from generative AI sources will spike 3,200% during Prime Day, as shoppers turn to AI tools to find deals and get a head start on back-to-school shopping.

• Amazon pulls AI-made political bios

  The books contained false claims about Scottish leaders, highlighting AI’s growing role in election misinformation.

• Nvidia’s sovereign AI tailwind gains momentum

  Citi raised its NVDA price target to $190, citing rising demand from state-backed AI infrastructure projects. They now estimate a $563B AI compute TAM by 2028.

• U.S. and Israel sign AI-energy innovation pact

  Interior and Energy officials inked an MOU with Prime Minister Netanyahu to co-develop AI tools for grid optimization, data-center demand, and cybersecurity, marking a new bilateral push to harden and modernize energy infrastructure.

• Video game actors vote on new AI-protection contract

  SAG-AFTRA’s tentative deal offers pay raises and consent rules for AI replicas, potentially ending an 11-month strike.

💰 The numbers

• Meta invested $3.5B in EssilorLuxottica (maker of Ray-Bans) to accelerate AI-powered smart glasses.

• LangChain, which builds dev tools for LLM apps, is raising at a $1B valuation led by IVP.

• Mistral is raising $1B to expand open-weight LLMs and build an AI superhub in Europe.

• Boldstart Ventures closed a $250M fund to back early-stage AI, cybersecurity, and crypto startups.

• PS Miner raised $350M Series B to scale its AI-powered cloud mining infrastructure.

• Heavybit raised $180M+ to fund dev-first startups building AI and infra tools.

• SiPearl raised a €130M ($152M) Series A to scale its AI + supercomputing chips for the EU’s Jupiter project; TSMC will manufacture the new 80-core Rhea1.

• Parspec raised a $20M Series A to automate construction procurement with AI, serving 288 customers and scaling across the $1T MEP supply chain.

🧠 Thought starters

• AI can now track how your cells age

• Scientists create biological 'artificial intelligence' system

• Anthropic’s store management experiment

This CFO Dive article got me thinking.

It’s true, most AI projects don’t deliver obvious ROI right away. And maybe they shouldn’t.

That’s why some CFOs are starting to use AI-adjusted ROI. Because it’s a more “honest” way to track impact.

The real value often shows up later as the system learns, speeds up decisions, and compounds over time.

We're entering a phase where AI agents won't just work on tasks. They'll measure their own impact and optimize the systems they're building.

Might be smart to start preparing for that.

😂 Meme of the day

Thanks for reading,

Eddie

Are we friends on LinkedIn and X? Let’s connect — DM me or hit the follow button.

P.S. If this was valuable, forward it to a friend. If you’re that smart friend, subscribe here.

P.P.S. Interested in reaching other ambitious readers like you? To become a sponsor, reply to this email.