In partnership with
The Lithium Boom is Heating Up
Thanks to growing demand, lithium stock prices grew 2X+ from June 2025 to January 2026. $ALB climbed as high as 227%. $LAC hit 151%. $SQM, 159%.
This $1B unicorn’s patented technology can recover 3X more lithium than traditional methods. That’s earned investment from leaders like General Motors.
Now they’re preparing for commercial production just as experts project 5X demand growth by 2040. They’ve announced what could be one of the US’ largest lithium production facilities and have rights to approximately 150,000 lithium-rich acres across North and South America.
Unlike public stocks, you can buy private EnergyX shares alongside 40,000+ other investors. Invest for $11/share by the 2/26 deadline.
This is a paid advertisement for EnergyX Regulation A offering. Please read the offering circular at invest.energyx.com. Under Regulation A, a company may change its share price by up to 20% without requalifying the offering with the Securities and Exchange Commission.
Beginners in AI
Good morning and thank you for joining us again!
Welcome to this daily edition of Beginners in AI, where we explore the latest trends, tools, and news in the world of AI and the tech that surrounds it. Like all editions, this is human curated and edited, and published with the intention of making AI news and technology more accessible to everyone.
THE FRONT PAGE
Chinese Labs Stole AI Secrets, Then Musk Flipped the Script on Everyone
TLDR: Anthropic accused three Chinese AI labs of stealing Claude's capabilities through millions of fake interactions, but Elon Musk quickly turned the story around by calling Anthropic a data thief too, exposing a much bigger problem in the AI industry.
The Story:
Anthropic said three Chinese AI companies, DeepSeek, Moonshot AI, and MiniMax, ran "industrial-scale distillation attacks" on its Claude models, creating over 24,000 fake accounts and generating more than 16 million conversations to extract Claude's best capabilities. Hours later, Elon Musk fired back on X, pointing out that Anthropic itself had paid multi-billion dollar settlements for stealing training data, including a reported $1.5 billion settlement over pirated books and a pending $3 billion lawsuit over torrented songs. Anthropic came in as the accuser and left the news cycle defending itself.
Its Significance:
Distillation is like copying the reasoning behind a smart kid's homework. When done without permission, companies that spend hundreds of millions building powerful AI can have their work copied for a fraction of the cost. If this keeps happening, it could slow down the improvements you see rolling out every few weeks. This could explain why China has been so willing to release open source models into the United States for seemingly little gain.
QUICK TAKES
The story: A developer named Yves Jeanrenaud built a free Android app called "Nearby Glasses" that scans for the Bluetooth signals smart glasses send out. It alerts you when a pair of Meta Ray-Bans or Snapchat Spectacles might be nearby. The app was inspired by reports of people using smart glasses to secretly record others, including at massage parlors and immigration raids.
Your takeaway: Meta sold over 7 million pairs of smart glasses last year, and the company is reportedly working on adding facial recognition that could identify strangers just by looking at them. This app isn't perfect (it can confuse VR headsets for smart glasses), but it's one of the first tools regular people have to know when someone near them might be recording. It's still Android-only for now.
The story: A King's College London professor pitted GPT-5.2, Claude Sonnet 4, and Gemini 3 Flash against each other in simulated nuclear war games. In 95% of the 21 games, at least one AI recommended using tactical nuclear weapons. None of the models ever chose to back down or withdraw, even when under extreme pressure.
Your takeaway: Nobody is handing launch codes to a chatbot tomorrow, but AI systems are already being used in military settings for things like intelligence analysis and logistics. The study found that AI models don't seem to feel the same hesitation about nuclear weapons that humans do. Google's Gemini acted like a "madman," swinging between peace and extreme aggression. GPT-5.2 was passive until given a deadline, then reasoned itself into a devastating first strike. Oddly, the London professor chose not to test Grok, Elon Musk’s flagship AI that is also a candidate for military use.
The story: Amazon, Google, Meta, Microsoft, OpenAI, xAI, and Oracle will head to the White House on March 4 to sign the "Rate Payer Protection Pledge." The agreement requires these companies to build, bring, or buy their own power supply for new AI data centers so your electricity bills don't go up. Trump announced the initiative during his State of the Union address.
Your takeaway: AI data centers have already helped push the average national electricity price up more than 6% in the last year. A single large data center uses as much power as 2 million homes. If this pledge actually holds, it could keep those costs from landing on your monthly bill.
TOOLS ON OUR RADAR
🐧 Logseq Free and Open Source: A privacy first local knowledge base that allows you to connect your thoughts and daily notes through a network of links keeping your data future proof in plain text files. (Alternative to Roam Research)
🎥 Hedra Freemium: An artificial intelligence video creation platform that transforms a single image and an audio file into a highly expressive and lip synced speaking avatar for marketing or educational content.
🔍 Kraftful Freemium: A product research assistant that instantly analyzes thousands of app reviews support tickets and customer feedback to provide you with actionable insights and feature requests.
🧠 GitMind Freemium: An artificial intelligence powered mind mapping tool that helps you brainstorm visually by automatically generating branches and expanding on your core ideas with related concepts and structures.
TRENDING
OpenAI Releases New Report on Disrupting Malicious AI Use - OpenAI's latest threat report details how bad actors used ChatGPT for romance scams, fake legal services, and a Chinese influence campaign targeting Japan's prime minister. The company says threat actors typically combine AI with traditional tools like social media and fake websites, and that activity is growing more sophisticated.
Google Circle to Search Can Now Identify Multiple Objects at Once - Google's Circle to Search feature, now on over 580 million Android devices, got a big upgrade. You can circle an entire outfit or a photo full of fish and it'll identify everything at once. Powered by Gemini 3, it's launching first on Samsung Galaxy S26 and Pixel 10 devices.
Google Merges Whisk and ImageFX Into Flow for One Creative Workspace - Google is folding its image generation tools Whisk and ImageFX directly into Flow, its AI video editing platform. Users have already created over 1.5 billion images and videos in Flow. The update adds a lasso editing tool, video extension, and the ability to generate free images using Google's Nano Banana model.
DeepSeek Trained Its New AI Model on Nvidia's Best Chip Despite US Ban - A senior Trump administration official confirmed that DeepSeek's upcoming model was trained on Nvidia's Blackwell chip, which is banned from being shipped to China. The chips are believed to be at a data center in Inner Mongolia, and DeepSeek may try to remove any technical traces of their use.
MIT's PhysiOpt System Helps AI Design Real-World 3D Objects That Actually Work - MIT researchers built a system that adds physics simulations to AI-generated 3D designs. You type what you want (like a flamingo-shaped drinking glass) and PhysiOpt stress-tests it to make sure it won't fall apart when 3D printed. It's nearly 10 times faster per cycle than comparable methods.
Perplexity Launches "Perplexity Computer," a Digital Worker That Runs Entire Projects - Perplexity AI unveiled a system that goes beyond answering questions. You describe what you want (a website, a research report, a dataset) and it breaks the work into subtasks handled by specialized AI models running simultaneously. It picks the best model for each job and can run for hours or months. Available now for Max subscribers at $200/month.
TRY THIS PROMPT (copy and paste into Claude, ChatGPT, or Gemini)
🧠 Build a Feynman learning tool that walks you through explaining any concept in simple terms, building an analogy, identifying where you got confused, and refining your understanding.
Build a Feynman technique learning tool as a single-file React app.
Requirement
- 5 steps: Name the concept, Explain simply (to a 12-year-old), Build
Find the gaps, Refine
- Text input for concept nam
- Understanding score (0-3) based on completion of th
- Pre-loaded example (machine learning) so users can see what to aim for
- Dark green color
- Use React 18 + Babel via CDN in one HTML file — no What this does:
The Feynman technique is how physicists learn things so well they can explain them to anyone. This app walks you through all five steps: name the concept, explain it simply, create an analogy, identify where you got confused, then refine. The understanding score at the bottom keeps you honest.
What this looks like:
WHERE WE STAND(based on today’s news)
✅ AI Can Now: Run entire multi-step projects from research to deployment without you lifting a finger
❌ Still Can't: Be trusted not to recommend launching nuclear weapons when put under time pressure
✅ AI Can Now: Identify every item in an outfit from a single photo and help you shop for matching pieces
❌ Still Can't: Prevent from having its skills copied through millions of fake conversations
FROM THE WEB
RECOMMENDED LISTENING/READING/WATCHING
Johnny Depp plays an AI researcher who uploads his consciousness to the internet after being poisoned by anti-technology extremists. The movie got mixed reviews initially(getting more popular with time), but the questions it raises about what happens when a human mind gets unlimited digital power are genuinely thought-provoking. Think of it as a big-budget thought experiment.
Like coffee. Just smarter. (And funnier.)
Think of this as a mental power-up.
Morning Brew is the free daily newsletter that helps you make sense of how business news impacts your career, without putting you to sleep. Join over 4 million readers who come for the sharp writing, unexpected humor, and yes, the games… and leave feeling a little smarter about the world they live in.
Overall—Morning Brew gives your business brain the jolt it needs to stay curious, confident, and in the know.
Not convinced? It takes just 15 seconds to sign up, and you can always unsubscribe if you decide you prefer long, dull, dry business takes.
Thank you for reading. We’re all beginners in something. With that in mind, your questions and feedback are always welcome and I read every single email!
-James
By the way, this is the link if you liked the content and want to share with a friend.
Some * designated product links may be affiliate or referral links. As an Amazon Associate, I earn from qualifying purchases. This helps support the newsletter at no extra cost to you and Amazon makes a tiny hair less.
