The Ethereum not ETH stuff is the mental fallacy that triggered me into writing and podcasting in the first place.
There is no strong Ethereum without an ETH worth trillions. Without ETH as a global store of value, Ethereum is a failed project. Full stop.
ETH is economic bandwidth for DeFi. It is the only asset maximized for CROPs, fail at high value ETH, fail at CROPs, fail at Ethereum.
Saying you’re bullish Ethereum not ETH is like saying you’re bullish America not the American economy. They are one and the same - economic engines.
Better to admit Ethereum is a failed project than “Ethereum not ETH”.
So spew that weak blockchain not crypto stuff out of your mouth, it doesn’t make sense for BTC, ZEC, ETH, or any truly crypto native project.
Aztec Alpha went live last quarter. The first privacy ZK rollup with full private smart contract execution on Ethereum mainnet.
We shipped a multi-asset fee payment contract (FPC) to testnet. FPCs are native to @aztecnetwork. What was missing was a version that supports non-native assets. With ours, users bridge USDC and pay fees in it directly, with no separate fee token to acquire first. One less wall between a new user and their first transaction.
Our Aztec Governance Dashboard went live. Our AztecNodes explorer added mainnet support. Plus developer tools: a faucet, a linter, an installable set of agent skills. And underneath, the protocol kept getting harder to break. 34 audit findings closed. Incident response through three post-launch events.
Privacy on Ethereum was a research direction for a decade. Last quarter, it became a working stack.
I'm starting to see AI being appropriately measured as a counter party risk.
AI is valuable and useful, although to the degree that you place your engineering skill in the hands of one party, is to the degree that you accept counter party risk.
I'm finding that I generally use AI to increase output for generalized tasks, or as a secondary opinion for contrast against my own.
Although, very very clearly, systems thinking and understanding is more valuable than ever.
Exhibit a:
https://t.co/Yfoi6LGUXt
I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.
As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!
I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.
It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?
88ms => 1.5ms
150K allocs => ~500 allocs
Incredible right? Nope.
My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.
This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.
The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.
Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.
Prove what's required. Reveal nothing else.
ZKPassport is joining @AztecLabs_, but the mission remains the same. 130+ countries, zero data uploaded, zero data stored.
Privacy that works in the real world: https://t.co/gFhFy3dHX7
News to share: the Blocknative team has joined Deloitte. It's been eight-plus years. Thousands (more?) of conversations. Too many talks, panels, blog posts, and podcast appearances to track. A team that refused to ship anything we weren't proud of. Grateful for each and every one of you. Onward. 🫡
Ethereum is fixing its 5 biggest issues this year.
Glamsterdam (H1 2026):
> EIP-7732: Block building moves on-chain, external relays no longer control what enters a block.
> EIP-7928: Parallel execution, transactions run simultaneously, L1 targets 10,000 TPS.
> EIP-7623: Smart contract calls get 78% cheaper.
Hegota (H2 2026):
> Verkle Trees cut node storage by 90%.
> A laptop is enough to run a node.
> Lower hardware requirements = more validators = deeper decentralization.
Speed is easy to copy.
Decentralization at scale is not.
With the advent of AI, it is very seductive to outsource your thinking or the work of understanding. You have to do the mental effort, to have quality output.
If you don't know the difference between quality or slop, you are making slop.