Look at this chart. Dot Com bubble + AI bubble. Same pattern. Same greed. Same inevitable pop. IMO Greed is deeply entangled with human nature.
It doesn't disappear after a crash. It just redirects.
When this stock market AI FOMO pops and it will pop people won't suddenly become responsible investors.
They'll look for the next speculative playground to multiply their money.
Every single time G
2000: Dot Com crashes
2001-2003: Gold and commodities pump
2008: Housing crashes
2009-2017: Bitcoin born and explodes
2020: COVID crashes everything
2021: Biggest altseason in history
The pattern is right there in front of us. After every crisis comes the rotation into higher risk assets. Because greed doesn't die. It migrates. When TradFi burns them where do you think that desperate money flows?
Crypto Bro! Every. Single. Fucking. Time.
The $725 B #AI infrastructure bubble is already showing cracks. Microsoft telling 100K engineers to stop using Claude. Uber burning annual budgets in 4 months.
When it pops the greed doesn't vanish. It finds a new home.
And crypto is the most speculative asymmetric bet on the planet.
RWA. AI protocols. DeSci. DeFi. All waiting for that rotation.
👽 BREAKING BRONDOR - VITALIK FIGHTS TO SAVE $ETH FROM FUD - HERE'S MY HONEST TAKE
8 senior contributors leaving in 2026. 5 in May alone.
Tim Beiko gone. Barnabé Monnot gone. Protocol Cluster decimated. CT screaming #Ethereum is dead.
And you know what? I'm bullish G
Here's why everyone is reading this wrong.
Vitalik just said the Ethereum Foundation will be a "smaller ship" that sells LESS ETH and focuses only on what matters: censorship resistance, openness, privacy and security. CROPS.
He literally said "being as fast and as scalable as possible, and only a small epsilon more decentralized than the others, is a route to mediocrity. If we try it we will lose."
Read that again.
Vitalik is NOT trying to compete with Solana on speed. He's playing a completely different game.
While everyone screams about TPS and transaction fees Vitalik is focused on making Ethereum provably bug-free using AI-assisted verification.
Something considered impossible until 6 months ago.
The departures?
Senior people leaving to build their OWN projects in the ecosystem. Dankrad Feist left to raise $1 BILLION for a new Ethereum advocacy organization.
The EF holds only 0.16% of all ETH. Compare that to other foundations holding 10-50%.
Ethereum is already the most decentralized smart contract platform by far.
IMO this FUD is the best entry you'll get G
Ethereum has: BlackRock BUIDL fund + Most #RWA tokenization + Most #DeFi TVL + Most developers ETF approved and accumulating
And CT is bearish because some researchers left to start their own companies? 💀
The weak hands sell the FUD. The smart money accumulates the fundamentals.
$ETH isn't going anywhere. It's just getting leaner and meaner.
👽 ALTCOIN SEASON 2021 MADE A LOT OF MULTI MILLIONAIRES
But few of them kept the wealth. Me included kek...
Here's what I did wrong and why I'm full into $QNT and $TAO ecosystem now
2021 MISTAKES
I held too long. Watched millions turn to thousands thinking "it will go higher"
I had no exit strategy. No targets. Just vibes and hopium.
I was in narratives not infrastructure. Hype coins with no real use case. I didn't understand what I owned. Just followed CT calls blindly.
WHAT CHANGED
This cycle I'm positioned differently.
$QNT - The bridge connecting TradFi to blockchain. SWIFT integration. ECB digital euro pilots. Banks actually using it. Infrastructure not hype.
$TAO - The Bitcoin of AI. Decentralized intelligence network. But here's where it gets interesting G
#TAO SUBNETS
Bittensor isn't just one protocol. It's 128+ subnets each doing different AI tasks.
I'm longing decentralized GPU and GameFi subnets without writing a single line of code.
Why? Because compute demand is exploding.
Microsoft just told 100K engineers to stop using Claude because bills exploded.
AI costs more than humans now!! Decentralized compute is the answer. And TAO subnets are building it.
GameFi subnets tapping into a $282B industry that's larger than movies and music combined.
Real revenue. Real users. Real infrastructure.
THE DIFFERENCE
2021 me: chasing pumps hoping for the best
2026 me: positioned in infrastructure that survives bear markets and explodes in bulls
I'm not making the same mistakes twice.
This time I have exit targets. This time I understand what I own. This time I'm in protocols that generate actual value. $QNT and $TAO are conviction plays.
I've seen too many great protocols fail due to poor money management.
There was a time, about two years ago, when low-cap protocols were spending 15k on a single PR on a Web3 blog, or 3-6k on a single tweet from major X accounts.
This era is over, but many protocols died as a result of this learning process.
The next bull market will be fairer for end consumers (retail investors) and far less rewarding for larpers and shillers who have no brains.
As always, I choose quality.
I've never accepted money from a gambling company (even if they left 100k or more on the table, my soul is not for sale).
I stick with quality enterprise protocols and real innovators.
That's it.
If you want to know about some hard-working people in the space that I'm still keeping an eye on, check these out:
Đây là quan điểm của tôi và lý do tại sao $TAO đang sẵn sàng cho cuộc chạy đua đến sự giàu có thế hệ G
Không phải lời khuyên tài chính đâu, nhưng cho tôi phân tích những gì đang xảy ra
#Microsoft vừa yêu cầu 100.000 kỹ sư NGỪNG sử dụng Claude vì chi phí đã bùng nổ.
Công ty đã đầu tư 5 TỶ ĐÔ vào #Anthropic giờ đang buộc mọi người phải rời khỏi Claude Code trước tháng 6. Quá đắt.
Uber đã đốt cháy toàn bộ ngân sách AI hàng năm của mình trước tháng 4.
Những người dùng nặng đang tiêu tốn từ 500 đến 2.000 đô la mỗi tháng MỖI NGƯỜI.
CTO đã chi 1.200 đô la trong một buổi demo kéo dài hai giờ.
Sau đó, phó chủ tịch của NVIDIA đã nói rõ: "chi phí tính toán cao hơn nhiều so với chi phí của nhân viên"
725 TỶ ĐÔ trong chi tiêu hạ tầng AI năm nay và những công ty đầu tiên triển khai quy mô đã BẮT ĐẦU rút lui vì hóa đơn đến trước khi có sự gia tăng năng suất.
Giá cả AI tập trung đang gặp vấn đề.
Kinh tế không hoạt động ở quy mô lớn.
OpenAI tính phí tùy ý.
Anthropic cũng tính phí tùy ý.
Microsoft chi 500 triệu đô la hàng năm mà vẫn không thể duy trì.
Vậy điều gì sẽ xảy ra khi các doanh nghiệp nhận ra họ không thể chi trả điều này mãi mãi?
Họ sẽ tìm kiếm các phương án thay thế.
Và $TAO đang xây dựng chính điều đó. Tính toán phi tập trung nơi những người khai thác cạnh tranh để cung cấp đầu ra AI.
Không có công ty nào đơn lẻ thiết lập giá cả. Động lực thị trường xác định chi phí. Cạnh tranh thực sự thúc đẩy hiệu quả.
Các thợ mỏ TAO đã tạo ra 43 triệu đô la trong Q1 2026 từ các dịch vụ AI thực sự. Không phải hứa hẹn. Doanh thu.
Cuộc khủng hoảng chi phí AI tập trung thực sự là chất xúc tác cho việc áp dụng AI phi tập trung. Khi bong bóng vỡ cho OpenAI và Anthropic, hạ tầng nào tồn tại sẽ là hạ tầng không phụ thuộc vào sức mạnh định giá của các tập đoàn.
Tôi không nói điều này để bơm túi của mình (tôi thực sự không phải ai cả!)
Nhưng tôi đang nói điều này vì toán học đang ngay trước mắt chúng ta.
AI tập trung đang tự định giá mình ra khỏi sự tồn tại. AI phi tập trung là bước đi hợp lý tiếp theo.
👽 $QNT and $TAO The two protocols I’m betting my future on.
(NFA: don't follow my lead; it's only for information).
Different niches. Same energy.
Both are building the infrastructure that the world needs but doesn't yet understand.
QUANT = THE BRIDGE
TradFi is a $400 trillion machine running on legacy systems. Banks can’t communicate with blockchains.
Blockchains can’t communicate with each other.
Quant has built the universal translator.
Overledger connects everything without forcing anyone to rebuild.
SWIFT has chosen them for cross-border testing. The ECB is running the digital euro on their technology. HSBC and Barclays are using it too. ISO standards have been published.
While the CT debates memes, Quant is becoming the backbone of institutional finance.
BITTENSOR = THE BRAIN
AI is a $15 trillion opportunity controlled by three companies. OpenAI can ban you. Google and Anthropic can change their terms overnight.
TAO has built an escape route. It is a decentralised AI where models compete and get paid based on quality.
The maximum supply is 21 million, the same as Bitcoin. Miners produce actual intelligence, not just securing a ledger.
Grayscale has filed an ETF application. Jensen Huang is discussing it publicly. $43M revenue in Q1.
THE CONNECTION:
QNT is the base layer for institutional blockchain connectivity.
TAO is the base layer for decentralised AI.
Both are solving problems that will only become more significant.
Finance needs connectivity. AI needs decentralisation.
These two are the answers hiding in plain sight.
Not financial advice, but you know where I stand, Tribe.