What happened with Galgotias University is proof we are producing degree holders, not educated humans.
Quality education is not about placements, itโs about character. If a university canโt teach responsibility, respect and basic civic senseโฆ itโs just selling certificates.
This is the only tweet you will need to survive the AI era.
10 rules. No fluff.
Read it carefully.
1. Be fast. The #1 skill in the AI era is high agency. Information is no longer a moat - you can prompt AI and find out anything in seconds. Your edge is speed. How fast can you spot an opportunity, act on it, fail, learn, and go again? While most people are still just reading about AI, a small % are already building with it. The gap between those two groups gets wider every single day. Don't overthink. Don't wait for perfect conditions. Move, test, learn, repeat. Speed is the new intelligence.
2. Build a business. Not many talk about it, but this is by far the best way to capitalise. I am personally focused on consumer products/media. There is still a massive arbitrage between the tech and the market. The gap is closing but it's still there. Playbook: Hire global talent before everyone else catches on (i.e. an EA in the Philippines costs ~$800/mo). Stack a small team around you, use AI to 10x their output, and you now run a machine that prints leverage. The gap between "solo operator with AI + a team" and "traditional company with 50 employees" is closing fast. Be on the right side of that. If you don't know how, move on to the next point.
3. Use AI EVERY single day. Use it to automate the boring stuff. SOPs, research, first drafts, data analysis, scheduling, and email. Anything repetitive that you do on a daily/weekly basis. Free up your brain for the stuff AI can't do yet - taste, judgment, relationships, decisions under uncertainty. People who use AI to replace their own thinking are getting dumber. Instead, use it to automate workflows and bring ideas to life. I post plenty of content on how to do this.
4. Understand that EQ is the new IQ. When everyone has access to the same AI tools, the differentiator is no longer intelligence, it's how you make people feel. How you read a room. How you negotiate. How you lead when things get messy. The most valuable skill in an AI-saturated world is being a human that other humans actually want to work with. This can't yet be automated.
5. Your health is your biggest edge. When everyone has access to the same AI tools, the person who sleeps 8 hours, trains hard, and eats clean will outperform the genius running on 5 hours and caffeine. Energy, clarity, and consistency are the new unfair advantages. Optimise sleep above everything - the work sorts itself out when your brain is firing on all cylinders. Side note: AI can replace a lot of things, but it can't replace your health. It's the one thing that you truly own, and that will never change. Respect it.
6. Build a personal brand. Now. AI can generate content, build products, write code. but it can't be YOU. Your story, your taste, your track record, your reputation - that's the one thing that compounds and can't be cloned. I was a 20-year-old kid tweeting from my parents' house in Australia. That decision to build in public changed my entire life. You don't need to be an influencer. You just need to be visible. Distribution IS the moat once barrier to code is no longer a factor.
7. Learn to sell. This is the skill no one talks about. AI can build the product, write the copy, design the site, and run the ads. It still can't shake a hand, read hesitation in someone's voice, or close a deal over dinner. Sales and marketing are the last pure human skills โ and the most valuable ones in a world where building is no longer the bottleneck. Distribution is. If you can sell, you will never be replaceable.
8. Invest in the picks and shovels. Aside from investing in your own tools, education and business, invest in the broader AI economy. Think: energy (AI needs insane amounts of power), infrastructure, compute, robotics, cybersecurity, biotech, blockchain, hard assets. Work out where the world will be in 10-15 years and reverse engineer your capital allocation today to reflect that.
9. Network and surround yourself with good energy. Stop hanging around people who complain about AI taking jobs. Start spending time with people who are using it to create them. Your network in the next 12 months will determine your net worth for the next 12 years. This isn't motivational fluff - the collaborations, ideas, and deals that come from being in the right rooms compound faster than anything else. Networking and human interaction is also something that AI can't replace. It might be the most powerful thing the machines can't take from us.
10. Stay curious. This is the meta-skill that feeds everything above. The AI landscape is shifting weekly. The people who win won't be the ones who "figured it out" once - they'll be the ones who never stopped learning. Read, explore, go down rabbit holes, try new tools, break things. I've been experimenting with all kinds of workflows, OpenClaw, MCPs, automations etc. If you want to see what I'm working on give me a follow, I share a lot of stuff for free.
If you are struggling to keep up with the fast moving AI industry,
here is the shortest roadmap that does the bare minimum in 2 weeks and actually works.
No hype. No rabbit holes. Just enough to stop feeling lost.
Week 1. Build the mental map
Day 1. What AI actually is today
AI today = Large Language Models + tools around them.
Understand what an LLM is and why transformers matter.
https://t.co/lehoO8UMm3
https://t.co/1qyNoW2X8Q
Day 2. How models are trained
Only learn the pipeline.
Data โ pretraining โ fine tuning โ inference.
Ignore math. Focus on cost and scale.
https://t.co/WfHSwrS9xT
Day 3. The big model families
Know who builds what and why people choose them.
GPT, Claude, Gemini, LLaMA, Mistral.
https://t.co/EZKBifZB9d
Day 4. Prompting that actually matters
Forget fancy prompts. Learn only this.
Context. Constraints. Examples.
https://t.co/mno8aO5iax
Day 5. Tools and agents
Understand function calling and agent loops.
Most agents are just prompts plus retries.
https://t.co/zvvUiCcW4S
https://t.co/iVpOVf8Dli
โธป
Week 2. Become practically dangerous
Day 6. APIs at a high level
Know what an API call looks like, what tokens cost, and why latency matters.
https://t.co/zikyBUahpk
https://t.co/Z0r1W5rkiT
Day 7. Retrieval Augmented Generation
LLMs + your data โ training.
Understand embeddings and vector search.
https://t.co/myoirwfe65
Day 8. Local vs hosted models
Learn when people say run locally, on device, or edge AI.
https://t.co/sYdzVTyir2
Day 9. What breaks in production
This is where real engineers live.
Hallucinations, cost explosions, latency spikes.
https://t.co/LSK8dz0kiM
Day 10. The AI product layer
AI features are not AI products.
Most startups die here.
https://t.co/WyJ9l2uKg7
Day 11. Job impact
Ignore doomsday takes. Look at workflow changes.
https://t.co/jywxWOsgv6
Day 12. Read one serious blog
Pick one and go deep.
https://t.co/CXwl5e6osg
https://t.co/AphoXogFoU
https://t.co/nKp8wPpo6q
Day 13. Build one tiny thing
A prompt workflow, internal tool, or small automation.
Building collapses confusion.
https://t.co/djTOOPYCkY
Day 14. Synthesize
Write one page.
What AI does well.
What it fails at.
Where cost and latency show up.
Where you personally can use it.
โธป
You do not need to chase every model release.
You need a stable mental model and light hands on exposure.
Two weeks of this puts you ahead of most people posting about AI.
Save this. Bookmark it. Come back to it.
How to value Gold & Silver
When we presented this data nearly 2 years ago and again last year, Gold & Silver were at a 50% and 60% discount to the midpoint of theoretical price framework. At the time it looked ridiculous. Then they rallied hard.
It once again looks silly given the momentum in prices.
But it is always good to revisit frameworks that hold the test of time.
Today's both Gold & Silver are trading above the top range of this framework. Tread with caution.
This framework can never tell you how to time the precious metals.