In partnership with

Sign Up - Linkedin

Today’s lineup: AI training startups are flooded with cash, ChatGPT may be emerging as an early mental health warning system, and the global race for rare earth minerals is heating up. Let’s dive in—and stay curious.

• Mercor Hits $10 Billion Valuation Training AI Models

• AI Tools

• AI as an Early Warning System for Mental Health

• AI Guides

• The Rare Earths Boom

Get 50% off forever

📰 AI News and Trends

• Anthropic gets ~80% of its revenue from corporate customers, and its Claude language models have been praised for their aptitude in coding. Anthropic is surprisingly close to OpenAI in revenue. It expects to get to a $9 billion annual run, a big lead in revenue per user over OpenAI.

• Merge Labs, a brain-computer interface (BCI) startup, founded by Sam Altman, has hired an award-winning biomolecular engineer and may want to link brains to ChatGPT.

• Saudi Arabia is pouring money into data centers and working with U.S. and Chinese tech giants, landing its A.I. ambitions in the middle of a geopolitical tussle for tech power.

• Anthropic is continuing its push into financial services by launching Claude for Excel, which lets users analyze and modify spreadsheets by chatting with Claude in a sidebar

• Qualcomm announces new AI chips in data center push

Other Tech News

• Breaking News: Amazon to hire ~250k seasonal workers

  Breaking News: Amazon has ~500,000 jobs on the chopping block due to automation in the next few years - ~30k of them corporate jobs

  “We’re living in a world where the working bees are essential, but the hive no longer needs so many queens.” Amazon Prepares To Cut As Many As 30,000 Corporate Jobs

• Japan’s Prime Minister Sanae Takaichi, who took office this month as her country’s first female leader, will greet the US president on his visit to Tokyo to strengthen relations by promising to purchase US-made trucks and secure access to rare earths, which Japan produces around 90% of the critical minerals.

• The world has overshot its target of limiting global warming to 1.5°C and must now prepare for the “devastating consequences” of climate change, the UN secretary general warned. We may witness the devastation of Jamaica within a few days as a consequence of Hurricane Melissa.

• Grokipedia crashed shortly after it launched yesterday. It was live late evening with nearly 900,000 articles.

• Citi partnered with Coinbase to develop digital asset payment solutions for institutional clients, initially focusing on streamlining fiat on/off-ramps and payment orchestration with future plans for 24/7 fiat-to-stablecoin options.

• https://www.fastcompany.com/91427749/gen-x-spends-more-money-in-stores-and-online-than-other-generations-why-dont-retailers-notice

Final Days to Invest: Every City’s a Potential Surf Destination

Topgolf revolutionized golf by turning it into a social, tech-driven game for anyone. And they’ve made billions in annual revenue doing it. Surf Lakes is applying that same model to surfing, and investors can still join them until 10/30 at 11:59 PM PT.

Their patented tech creates 2,000 ocean-quality rides per hour, anywhere in the world, across all skill levels.

It’s a $65B surf tourism market opportunity. They’ve already sold 8 tech licenses, plus their own commercial parks are underway.

3x world champ Tom Curren and surf icon Mark Occhilupo have joined as ambassadors and shareholders. Even actor Chris Hemsworth has praised Surf Lakes.

You have until October 30th at 11:59 PM PT to invest in Surf Lakes.

Invest In Surf Lakes

_{This is a paid advertisement for Surf Lakes’ Regulation CF offering. Please read the offering circular at https://invest.surflakes.com}

Mercor Hits $10 Billion Valuation Training AI Models

Founded in 2023 by three Thiel Fellows, Mercor trains AI models for OpenAI and Anthropic using a global network of 30,000 contractors. The startup just raised $350 million, boosting its valuation to $10 billion, 5x higher than in February.

After Meta’s $14B investment in Scale AI, Mercor’s revenue quadrupled as clients sought alternatives. Contractors earn an average of $85/hour, with Mercor taking 30–35% in fees. Now the company faces a lawsuit from Scale AI, underscoring the fierce competition in the booming AI data-labeling industry.

What is AI Systems Training?

Training AI systems means feeding machine learning models human-labeled data and expert feedback so they learn to think, reason, and communicate like people. This involves tagging images, text, and audio, ranking chatbot responses (known as reinforcement learning from human feedback), and having professionals like doctors or lawyers review model outputs for accuracy.

Major players include Scale AI, Sama, Appen, Remotasks, and Mercor, which work with companies like OpenAI, Anthropic, and Meta. The global AI training data market is projected to exceed $30 billion by 2030 with a 25%+ CAGR, driven by demand for higher-quality and domain-specific data.

There’s still a significant opportunity for startups that specialize in niche data labeling (medical, legal, multilingual, or regional) or that build human-in-the-loop tools to make AI feedback faster and more accurate. In short, training AI remains one of the most critical and profitable layers of the AI ecosystem.

Leave a comment

What we are reading:

• Learn How To Achieve Financial Freedom With a Business Model Called Instagram Automation

• Join 280,000+ entrepreneurs as we dive into wealth-building strategies, cash flow management, tax strategy, investing, and lifestyle design.

Share Yaro on AI and Tech Trends | Your Top AI Newsletter

AI as an Early Warning System for Mental Health

OpenAI reports that about 560,000 ChatGPT users weekly show signs of mania or psychosis, and 2.4 million express suicidal thoughts or emotional dependence. With input from over 170 psychiatrists, GPT-5 can now detect distress and guide users toward help, reducing harmful responses by up to 52%.

Used responsibly, ChatGPT could act as an early-warning tool, spotting users in crisis and prompting professional intervention. The key is consent and human oversight: users would opt in to share emergency contacts, and trained reviewers would confirm any alert before notifying family or emergency services.

AI won’t replace care, but it could save lives by recognizing when someone needs help and ensuring that help arrives.

Share Yaro on AI and Tech Trends | Your Top AI Newsletter

🧰 AI Tools of The Day

Turn RAG into riches with Delve’s AI for security questionnaires

Hate those endless security questionnaires? We do too. So we hired the sharpest AI brains from Stanford, MIT, and beyond—and made them do your grunt work. Delve uses state-of-the-art agentic RAG, not your average “retrieve + answer” hack.

Our agents think ahead: pull evidence, resolve conflicts, reason across your policy graph, interrogate your infrastructure, and draft bullet-proof responses. We’ve helped Lovable, Bland, Micro1, and a ton of the fastest-growing AI companies navigate and close review with almost every F50 - and saved dozens of hours with our AI-native compliance platform.

Use YAROCELIS1KOFF for 1K off

The Rare Earths Boom

The U.S. and other thriving economies are pouring billions into rare-earth mining and processing to cut reliance on China, which still controls over 70% of global supply. After Beijing restricted exports this year, prices soared and factories stalled, sparking massive Western investment.

Major deals include Orion Resource Partners’ $1.8B fund, $2.2B in U.S. financing for Australian projects, and JPMorgan’s $10B national security initiative, starting with Perpetua Resources in Idaho.

Rare minerals like neodymium, dysprosium, gallium, and antimony, mostly found in China, Australia, the U.S., and Africa, are essential for EVs, chips, and defense tech. Companies like Lynas, MP Materials, and Iluka stand to benefit as nations race to secure supply chains that power the AI, green energy, and tech revolutions.

🧰 AI Guides