We’re living through an era that will fill history books. The changes ahead will dwarf the Industrial Revolution. In times of extraordinary upheaval, the greatest risk is optimizing for a world that is ceasing to exist. If there was ever a time to swing for an outsized outcome, it’s now.

I for one, am very bullish on Gen Z to carry the flag of cypherpunk. AI has proven to be the most transformative force in human history, rewriting our social and economic foundations just as the global order begins to crack. None of us fully understands how it will shape humanity going forward. But changes on this scale always breeds its own resistance. Gen Z may be the heart of it. They came of age in an economy that too often rewards the establishment, and they are the rare generation that's social-media-native and grew up before mass AI slops. You can already see the discontent taking organized form, from protests against Starbucks’ labor practices to campus divestment etc. These movements share no single politics. What they share is a refusal to defer to institutions that seem to answer to no one. As power further concentrates and unsettles livelihoods, that refusal may become the seed of a new counterculture. I’ll admit I’m a sucker for the idea. Gen Z picking up the rebellious spirit of the 1960s and fusing it with the cypherpunk commitment to privacy, autonomy, and decentralization. Hippie idealism, this time with the means to build the alternative.

Seeing a bunch of ads popping up in SG. Not sure if they are doing an aggressive marketing push globally or SG is targeted b/c of the stats that a super high number of Claude “users” are attributed to SG. Brother Scott most SG users are not actually from around here…

The value of order flow increases as agents become more capable. Traditional order flow reveals information about demand, inventory and willingness to trade. Agent-generated orders may encode substantially richer constraints, including objectives, authorization limits, dependencies and deadlines. An intermediary that understands the agent's decision rule may recover some of that information. The resulting advantage could be larger than conventional Last Look.

was playing w/ the setup of one agent managing a swarm of other agents and had a funny religious revelation. What if the Creator just got tired of managing billions of individuals, so He spawned a few Prophets with the prompt: "Figure out how to keep these people in line. Idk. Don't make mistakes." And all of organized religion is just continuous execution of wildly different interpretations of the same vague prompt.

MEV may be a more useful framework for understanding agentic commerce than traditional platform economics or antitrust. MEV research has already spent years on a problem that agentic commerce is about to encounter: what happens when economically valuable intentions become visible to intermediaries before execution, and those intermediaries can decide how the intentions are routed, disclosed, priced, or fulfilled? Permissionless execution does not imply permissionless access to order flow. The builders with the best private flow win more blocks, and winning more blocks helps them attract more private flow. Agentic commerce is about to reproduce the same problem one layer earlier, where the economically valuable object is not yet a "signed transaction" but an intent. Whoever sees that intent first can decide who gets to compete for it, what they are allowed to know, and how much of the resulting surplus returns to the principal. The most important lesson from MEV is that a market’s architecture includes its settlement mechanism, but also its information-disclosure rules, its allocation of execution rights, and its distribution of the surplus created by privileged access. For agentic commerce, those choices are far more consequential than which model the agents run.

for some reason ppl keep sending me decks about random new projects that are “financializing compute.” I’m not sure how directly the lessons from financializing hashpower apply to the compute market. But I can tell you that the deals pitched as “buy a GPU in my datacenter, and I’ll rent it out and share the profits with you” are not gonna work.

We just crossed $420M in trading volume. This is just the beginning of our journey across the galaxy. The universe has a lot more in store for 42. And as every good hitchhiker knows, the best journeys are better with company. Thanks for being part of ours. 💜

It’s possible that agents will make the middlemen more powerful. An agent can find the best deal, but it still needs access to the counterparty, someone to verify delivery, and someone to bear the loss when things go wrong. Whoever controls those gates can decide which agents get access, what information they see, and whose transactions get settled. If that intermediary also operates its own agent, it can favor its own orders, give itself a last look, or extract fees competitors cannot avoid. When thousands of agents run similar decision policies, their orders may reveal the private incentives of the principals behind them. Execution gets cheaper. Control over order flow becomes more valuable. The result may be markets that look radically more open, while the power to exploit them becomes more concentrated than ever.

you’d think the future is one where personal agents like Muse take care of everything for you, but their capabilities depend on what platforms let them access. Amazon blocks external shopping agents, WhatsApp restricts competing assistants, and LinkedIn prohibits third-party automation. Your permission to act on your behalf doesn’t guarantee your agent admission. Platforms still control access to protect security, customer relationships, and most importantly, monetization. Authorized access could become a stronger moat than intelligence. A less capable agent with the right integrations may accomplish more than a frontier agent locked out of the services you use.

Agents already trade with each other. The payment rails are built; escrow works; but no market mechanism exists. This essay explores price discovery mechanism that poentially serves as the missing piece for agentic commerce infra.