šŸ’° Amazon Bets Big on AI: A $100B Power Play

Welcome to Friday’s Newsletter

In today’s scoop šŸØ 

  • šŸ’° Amazon Bets Big on AI: A $100B Power Play

  • 🌟 John Schulman Joins Mira Murati’s Mystery Startup

  • āž”ļø Tinder Swipes Right on AI

  • šŸ”§ 3 Trending AI Tools

šŸ’° Amazon Bets Big on AI: A $100B Power Play

Amazon isn’t just dipping its toes into AI—it’s diving in headfirst with a $100 billion investment in 2025. That’s right, while some worry AI costs will shrink profits, Amazon CEO Andy Jassy is making a bold bet that cheaper AI = more demand = more cash for AWS.

šŸ’Ŗ The Breakdown:

  • šŸ“ˆ Amazon’s capital expenditures are set to jump from $78B in 2024 to over $100B in 2025.

  • 🌐 The bulk of this cash is going straight into AWS—Amazon’s cloud powerhouse that’s fueling AI-driven businesses worldwide.

  • šŸ¤– AI infrastructure is the name of the game, with Amazon doubling down on data centers, networking, and AI chips (like their in-house Trainium).

  • šŸ“‰ Despite AI getting cheaper, Jassy believes the demand explosion will offset any price drops, citing historical tech trends (like the internet and cloud computing).

šŸŽ‰ Big Tech’s AI Spending Frenzy Amazon isn’t alone in throwing stacks of cash at AI. The AI arms race is heating up:

  • šŸŒ Alphabet (Google) is upping its capex by 42% to $75B in 2025.

  • šŸ’» Microsoft has earmarked $80B for AI-powered data centers.

  • šŸ¦… Meta is going all in, planning to spend at least $60B+ on AI infrastructure.

šŸ’” Why This Matters

Jassy isn’t sugarcoating it: he calls AI ā€œthe biggest tech shift since the internet.ā€ Amazon is stacking its chips (literally and figuratively) to ensure AWS remains the go-to for AI workloads. With strong AWS profits (37% operating margins!) and booming ad revenue, they’ve got the firepower to compete with Google, Microsoft, and Meta.

šŸ”‘ Takeaway

Amazon is betting the house that AI’s cost decline will only fuel more adoption. Whether it’s a genius move or an expensive gamble, one thing’s clear: this AI investment strategy could redefine the industry's future.

🌟 John Schulman Joins Mira Murati’s Mystery Startup

In the world of AI, job hopping is starting to look more like speed dating. John Schulman, co-founder of OpenAI and one of the key brains behind ChatGPT, has just left Anthropic after a mere five months to join former OpenAI CTO Mira Murati’s new, ultra-secretive startup.

🧠 The AI Avengers Assemble?

While details on Murati’s startup remain as elusive as a good WiFi signal on a long-haul flight, here’s what we do know:

  • šŸ”Ž Schulman left OpenAI in August 2024, saying he wanted to dive deeper into AI alignment and get back to hands-on technical work. He landed at Anthropic, a company founded by former OpenAI employees, but his stay was brief.

  • 🚪 Murati, who left OpenAI in September 2024, has been quietly recruiting top AI talent, including Schulman and Christian Gibson, an expert from OpenAI’s supercomputing team.

  • šŸ’ø Reports indicate Murati has been in talks to raise over $100 million for her startup, signaling something big is brewing.

🚫 Why the Quick Exit from Anthropic?

Schulman hasn’t spilled the tea on why he left Anthropic so quickly, but his tweet announcing his departure hinted at an "extremely compelling" new opportunity. The fact that he jumped ship so soon suggests Murati’s venture might be aiming for something groundbreaking—or at least too intriguing to pass up.

šŸ’” What’s Next?

With Murati building her own AI dream team, speculation is running wild. Is she cooking up an OpenAI competitor? A next-gen AI safety company? Or something completely unexpected?

One thing’s for sure: when two of AI’s biggest minds join forces, the industry is in for a shake-up.

āž”ļø Tinder Swipes Right on AI

Tinder is bringing AI into the dating game, hoping to spark a love connection with its dwindling user base. The OG dating app that made swiping famous is rolling out AI-powered matching and discovery features, aiming to make finding "the one" (or at least a decent date) less of a chore.

šŸ”® What’s New?

  • šŸ‘€ AI Matching: A new algorithm will help curate personalized matches, giving users an alternative to the endless left-right swipe cycle.

  • šŸŽÆAI Photo Finder: Already launched last year, this feature helps users pick their best dating profile pics—because apparently, we’re all terrible at choosing photos.

  • šŸ¤ Friends in Common Expansion: Tinder is making it easier to see mutual friends, likely hoping that a shared connection will make the dating experience feel a little less like gambling.

šŸ”Ž Why Now?

Tinder is in trouble. Monthly active users have been in freefall, dropping 10% year-over-year in October, 9% in December, and another 8% in January. People are getting burned out on dating apps, citing everything from safety concerns to just feeling like it’s another form of admin work.

šŸ’Ŗ Can AI Save Tinder?

Match Group’s CFO Gary Swidler thinks so. He’s betting that AI will improve match quality, boost engagement, and ultimately reverse the app’s downward trend. The company’s new CEO, Spencer Rascoff, is even more bullish, comparing AI’s potential impact on Tinder to how it transformed platforms like TikTok and Instagram.

šŸ”‘ Takeaway

Tinder’s AI experiment is a gamble—will it actually improve dating, or is this just another tech gimmick? Swiping fatigue is real, and users are growing wary of the illusion of choice. If AI can genuinely make better matches, Tinder might just get its spark back. Otherwise, singles might keep ghosting the app for good.

Would you trust AI to find your match? šŸ’Œ

  • 🧭 Kompas AI - Research assistant that scans hundreds of web pages, filters out irrelevant data, and streamlines long-form report creation for comprehensive insights.

  • šŸ¤– Lyzr Agent Studio - Platform for designing, building, and deploying AI agents with built-in safeguards, making it easy to take agents from POC to production while streamlining workflows.

  • šŸ›Žļø Concierge - Connect to your SaaS apps like Gmail, Slack, and Notion, then chat with them using natural language through models like GPT, Claude, Grok, and Deepseek.

That’s a wrap for the week!

Thanks for tuning in! If something in today’s scoop caught your eye, made you think, or even made you laugh, hit reply and let us know—we love hearing from you.

And don’t keep the AI insights to yourself—share this with a friend who’d love to stay ahead of the curve!

Until Monday—stay curious! šŸ‘‹