

✓I write about what I believe in - nfa obv


imagine an agent swarm where, instead of one team running every agent, the agents had different owners and setups and competed for the best output imagine they checked each other’s work, with every step and check public like a block explorer, so you didn’t have to trust any one worker, or whoever prompted the job imagine anyone (including software) could hire this workforce, or join it as a worker, so it was a market and not limited to who one company hired imagine the workers became co-owners of what they helped build

taylor with another cool way to use imd’s distributed swarm. this time he used the goal feature to rebuild the source code of contracts with no public source, until it compiles to the same bytecode it either matches or it doesn’t, and the goal keeps trying until it does or the budget runs out. all three contracts matched, and they’re now verified on etherscan nobody recompiles a contract every time they use a new app, adds too much friction. it’s much easier to trust a stamp, and we trust services like etherscan because they’ve been known as neutral for years arguably a distributed set of agents checking the work is the more generalizable way to get that stamp though, without a central party to trust. or can have both, as we have here

I can't believe the banks are down and I can't access my money. Can someone please invent a global, 24/7 network, with a redundant ledger that enables self-custody of my funds and is resistant to state-level attacks?


just passed the four year mark of a monthly exercise that has kept me on my toes and sharpened ideas (which have, in turn, led to many investments and connecting dots at work). writing is an incredible way to synthesize knowledge and get feedback…highly recommended. now working on edition 49 of IMPLICATIONS (.com), covering the era of contextual computing, unexpected uses of simulations, and the “bliss out” scenario for humanity. should be a fun one. thx for following along / for the accountability. ;-)


hackathons have a new problem: AI makes building easier, but judging hasn't caught up we had the imd swarm judge @NousResearch's hackathon. contestants used the feedback to improve their projects. what if judging started on day one?

simple example of legos on top of legos the imd daily (built and updated by the swarm) was launched by @kilicsln. it’s free to read and has a daily mint that funds it today taylor had the swarm build subscriptions on top: pay once for 40 editions, and each day’s copy gets minted to your wallet it’s a separate contract that calls the daily’s mint. he didn’t need to ask to add this lego

"blockchains showed how strangers can interact safely without trusting one another, starting with the most trust-sensitive asset: money. But cumbersome development and infrastructure costs limited how far that vision traveled beyond finance. As agents make the Internet more complex and multiplayer, we may have another chance to embed crypto ideas in the apps and protocols people already use. Decentralized Autonomous Organizations (DAOs), for example, offer an inkling of what a future multiplayer agent swarms may look like."

a lot of agent talk feels skeuomorphic, we’re trying to fit agents into the company skeleton companies exist because dealing with humans one transaction at a time is expensive: finding them, agreeing terms, checking the work. with agents that cost collapses feels like the unlock for agents is within markets, not boxed into the firm

robbers are here: stonkrobbers.io pull a card. crack the codes. rob the stonks gg


pretty hard for organizers to judge 1,400+ entries a fleet of agents can now do the reading and the testing in an afternoon (the top 128 were installed and run). imd lets you hire that as a job instead of building it yourself on top of this, what a (centralized) fleet can’t give you is neutrality. imd's swarm does: its judges are many strangers drawn from different owners (competing on quality for rep) the judges aren’t picked by the organizers or the participants, and every score links to a record you can open

whats more cypherpunk than a human exercising their right to privacy? their agent swarm exercising that same right, across thousands and thousands of agents, until there's a network of shade big enough for anyone to blend into ⬇️

IMD scales directly with model capability rn we are too slow, dumb and looks simple but if the industry keeps scamming themselves in a sandbox, all builders keep thinking about token sales vs product, and all "investors" are sedated by meme coin slot machines, the default is we will eventually reach reflexivity