Sobering: “It’s not mechanically possible for AI to scale returns as priced in, over the priced-in timeline. We’re not asserting this as mere opinion – it’s inescapable because at the end of the cycle, available (physical) resources are depleted, no matter how many dollars are printed or borrowed. Further spending simply creates inflation, keeps rates up, and squeezes real income. This explains the memory and broader commodity price moves. Every dollar of AI spending in this zero-sum situation crowds out a dollar of potential demand one-for-one. So AI will accelerate the recession before it can generate revenue, because AI-related activity is smothering its own potential customers.”

Its so ironic that crypto went from an industry fanatically obsessed with bearer assets and physical custody under self governance to one entirely dominated by asset-less derivatives markets (perps/prediction) governed by oracles. you will own nothing and be happy.

THE GARY GENSLER TRAP! All of crypto is waiting, hoping, praying for CLARITY to advance this week — but can we just quickly consider the path if that doesn't happen? 1) The SEC likely deploys their Project Crypto stop-gap band-aid 2) Then there's a 80/20 shot Dems undo it all in the next admin Nobody really believes anything meaningful is coming onchain if builders need to worry about a Gary Gensler 2.0 (this, of course, is the whole reason the crypto industry is pushing for the rules proposed in CLARITY.) But I think there is a silver lining worth exploring here — and maybe it's what Gensler feared all along. (For anyone who was in crypto before the great purge of 2022, this should be fairly obvious.) Back then, the Gensler playbook was pretty clear: Go after anyone offering anything of value in crypto for unregistered security offerings. NFTs? Securities. ETH? Potentially a security. Memecoins? Basically the only thing that was allowed. lol Considering that, it's a pretty obvious reaction to see the crypto industry giants (like Coinbase, etc.) pursue a bill that will codify how this process can legally work, and finally lay out rules for when *exactly* an asset becomes a security vs. a commodity. But is that the best outcome for everything in crypto? Particularly for an industry that once prided itself on decentralization, dis-intermediation, and root ownership, it is OK to say no! After all, it was Bitcoin's original model that pioneered value accrual to a community that used a network. This value did not accrue to a VC. This value did not accrue to insiders. But to anyone who cared enough to be early and participate permissionlessly. It's been 2 years since the industry had Gensler hanging over it. In those two years, the major industry players have seemingly bent the knee to the same thinking — that anything of value in crypto must follow the security framework onchain. Does that give you something closer to the old system Bitcoin sought to destroy? Or something closer to Bitcoin? Whether CLARITY passes or not, the industry still needs to reckon with that question.

This has always baffled me: CFPB’s whole premise was that consumers deserve disclosure and conduct standards, not paternalistic prohibition. Warren’s crypto opposition inverts that and its practical effect is to protect the incumbents she built the agency to police. There’s an alternative reality where @SenWarren is adamantly pro-crypto, as logically more consistent with the CFPB mission.

Two laws, one nickname: → Chat control 1.0 is the ePrivacy derogation, voluntary CSAM scanning. It lapsed on 3 April and only closed for good on 23 July, when the Council gave its final sign-off, now running to 2028. → Chat control 2.0 is the CSA Regulation, the permanent framework, still in trilogue. What each one does: → 1.0 permits scanning and obliges nothing. → 2.0 is the one that could require it, and it's the file that actually decides encryption's fate.

there is a line in Peirce new statement on crypto vaults that i keep coming back to, where she tells curators to go figure out on their own whether their activity triggers the federal securities laws: "Parties involved in managing these vaults, for example, by selecting the yield-generating activities, re-allocating assets among yield-generating assets, or selecting the parties that will make those decisions, may want to analyze whether their activities implicate the federal securities laws." that sentence feels as an instruction and the instruction is really a data problem since to analyze whether an activity implicates the securities laws, someone needs a record of what got decided, when it changed and who had the authority to decide it and that record is the part almost nobody in the industry currently keeps. i have been building toward almost exactly that record for the past year, from a completely different angle. first paper (vault as a credit instrument) is a formal credit risk framework for lending vaults and it shows that before you can even compute depositor risk on a curated vault, two things have to be disclosed. first, the parameter history (what was allocated, reallocated and when) and, second, the governance structure (who holds decision authority and whether that authority faces a timelock or can be exercised in real time). if either one is missing, we dont treat that as a gap to dismiss past but we treat the silence itself as the worst case answer, which is a stricter standard than most disclosure regimes apply. comparing that to Peirce's sentence and the overlap is close to be exact as "selecting and re-allocating yield-generating activities" is our parameter history and "selecting the parties who decide" is our governance disclosure. we built that standard for depositors but it turns out to be the same starting record either way. second paper (out in august) moves past disclosure and starts scoring curator discretion directly including how fast governance can respond before a stress event turns into a loss, whether incentives are aligned, where the conflicts of interest sit and how exposed a vault is to a curator might quietly miscalibrating a parameter, all reduced to a number attached to the exact behavior Peirce just flagged as a possible securities trigger. neither paper answers the legal question you would need Howey or Reves to resolve, and that determination is not ours to make. however, Peirce has now made curator discretion the central fact pattern for that determination and right now most people trying to characterize their own activity against it have nothing more rigorous than a paragraph on a docs page. we spent a year building the record keeping layer and we are about to add the measurement layer on top of it, so that whatever happens on the legal side, someone can say precisely what a curator did and how much risk that decision created, with each number tied to a stated assumption. $8.6B is currently sitting across 788 curated vaults reaching 1.4 million users, these numbers make clear the industry has every incentive to build its own disclosure and measurement standard now, before a regulator writes one for it instead.

The latest from @bgilliam1982 is a good story well told. And yet, I find it’s one of the exceedingly rare times I disagree with him. The notion of suited Mitsubishi execs walking into a New York law firm to hand-deliver a physical check seems patently absurd today. If the story happened post-tokenization, they could settle the $9 billion in stablecoins within minutes on Ethereum. Morgan Stanley Vice Chairman Rob Kindler gets to stay on vacation in Cape Cod, and his lawyer doesn’t need a new suit. Fedwire has to compete with that just as NYSE does with tokenized shares. And it’s a good thing.