July 18, 2026 is the day the United States is supposed to have finished writing the rules for dollar stablecoins. It is one year to the day after the GENIUS Act, the first federal framework for payment stablecoins, was signed into law, and the statute set that one-year clock itself. The rules are not finished. As this is written every substantive rulemaking is still in proposed form, no final implementing rule has been published, and the two weeks around the deadline are when the search traffic, the headlines, and the lobbying all peak at once. This is the run-up explainer: what is actually due on July 18 and who owes it, what the Act's yield ban does and does not reach, what an issuer needs in hand before the rules land, and what happens if the deadline slips.
We have written the landscape already, in two companion pieces you can read for the fuller picture: the split between issuer-paid and strategy-passed-through yield, and the broader case for reserves you can verify after the depegs and the New York Fed's warning. This one is narrower and time-stamped to the run-up. It does not front-run rules that do not exist yet: every claim here is about the enacted statute or a dated, published proposal, and when the final rules actually land we will publish the rule-day analysis then, not before. Nothing here is legal advice, and the figures and status are as of the date at the top.
What is actually due on July 18, and who owes it
Start with the part most coverage blurs, which is who the deadline even binds. The Act's rulemaking mandate, Section 13(a), directs each primary Federal payment stablecoin regulator, the Secretary of the Treasury, and each State payment stablecoin regulator to promulgate implementing regulations not later than one year after enactment, which is July 18, 2026. The Act names four primary Federal payment stablecoin regulators, and which one governs a given issuer depends on what the issuer is: the Office of the Comptroller of the Currency, the Federal Reserve, the FDIC, and the National Credit Union Administration. Alongside them the Treasury Secretary owes a distinct set of rules, the anti-money-laundering, Bank Secrecy Act, and sanctions requirements, which in practice are being written by Treasury's own bureaus, FinCEN and the Office of Foreign Assets Control. Press shorthand sometimes counts six agencies, folding in FinCEN and OFAC as though they were separate from Treasury and, notably, leaving the Federal Reserve off the list. The precise version is the one worth holding: four named prudential regulators, the Treasury Secretary acting through FinCEN and OFAC, and the states, all on the same one-year clock.
Here is where that clock actually stands, with dates. The OCC published its proposed rule in the Federal Register on March 2, 2026, and its sixty-day comment window closed May 1. On April 10 two more proposals landed: a joint FinCEN and OFAC proposal on the anti-money-laundering and sanctions rules, and the FDIC's prudential proposal for the issuers it supervises, and both of those comment periods closed on June 9. The Federal Reserve, one of the four named primary regulators, had not published its own proposal as of late June. And as this run-up begins, no agency has published a final rule. Every one of these is still a proposal being turned into a final rule against the July 18 date.
One more date matters more than July 18 for anyone who actually has to comply, and it is not July 18. The Act's requirements take effect, under Section 20, on the earlier of January 18, 2027, which is eighteen months after enactment, or 120 days after the primary regulators issue their final rules. So the finish line for writing the rules and the starting line for obeying them are different lines. The 120-day compliance window everyone talks about only exists if final rules arrive in time for it to run before the January 2027 backstop. If the rules are late, there is no separate 120-day cushion; the eighteen-month date simply arrives.
What the yield ban actually says, and why it defines the whole question
The single line in the Act that reshapes the most is short. Section 4(a)(11), titled Prohibition on Interest, says that no permitted payment stablecoin issuer shall pay the holder of any payment stablecoin any form of interest or yield, whether in cash, tokens, or other consideration, solely in connection with the holding, use, or retention of such payment stablecoin. A compliant United States payment stablecoin cannot pay you for holding it. That is why the yield in this market has migrated into staking receipts and separate tokens: the coin itself stays silent, and the yield is paid on something adjacent to it.
The sharp edge of that line is what it does not reach, and it is the reason this is not only an issuer's problem. The prohibition binds issuers of payment stablecoins, which the Act defines as dollar tokens an issuer is obligated to redeem for a fixed amount. Derivatives-backed synthetic dollars, the delta-neutral designs like Ethena's USDe, are, on the prevailing legal reading, generally understood to fall outside that definition, because their issuer is not obligated to redeem them for a fixed sum. If a dollar is not a payment stablecoin, the yield ban does not attach to it, and neither do the Act's protections. A Forbes contributor put the same point in a June headline that has aged straight into the run-up: Ethena's USDe pays yield legally, and the GENIUS Act has no answer for it. We walked the mechanics of that split in the issuer-paid versus strategy-passed-through piece; the run-up point is only that the yield line is where compliant dollars and yield-bearing synthetic dollars formally part ways, and that the second category is left without the statute's floor. This is the prevailing interpretation of a contested gray area, not settled law, and the OCC's proposed rule reaches further than the statute, extending the prohibition toward yield paid through an affiliate or a related third party; the final rules may sharpen the edges either way.
What an issuer needs in hand before the rules land
Set aside what is still being drafted and a surprising amount is already fixed, because it lives in the statute rather than in the pending rules. An issuer of a compliant payment stablecoin will need, at minimum, all of the following operating before the requirements bite: reserves backing the coin at least one to one, held only in a short enumerated list of high-quality liquid assets, cash, short-dated Treasuries, government money-market funds and the like, and not rehypothecated; a monthly public disclosure of the reserve composition on the issuer's own website; a monthly examination of that disclosure by a registered public accounting firm; a certification of each monthly report by the issuer's chief executive and chief financial officer, carrying the same criminal liability as a false Sarbanes-Oxley certification; a published redemption policy with clear procedures and plain-language fees; a full Bank Secrecy Act program, because a permitted issuer is treated as a financial institution, with a sanctions-compliance program alongside it; the yield prohibition above; and marketing that does not imply government backing or insurance. That list is not waiting on the rulemaking. It is in Section 4 of the Act as passed. What the final rules still owe is the granular layer on top: the capital, liquidity, and risk-management standards, the tailored Treasury anti-money-laundering rules, and the application and supervision mechanics.
The large issuers are already staffing for all of it. In May, Falcon Finance and Anchorage Digital, a federally chartered bank, launched fUSD, a payments stablecoin marketed as GENIUS-ready with a monthly Deloitte attestation, sitting beside Falcon's larger and separate synthetic dollar USDf. A small issuer or a DeFi team cannot buy a Big Four relationship and a national bank charter, and for the statutory examination they should not try to fake one: that examination runs through licensed accounting firms and there is no software substitute for it. But there is one part of readiness that is not the examination and that a smaller issuer can stand up now, and it is the part a holder actually checks: reads of the disclosed reserve addresses that anyone can re-derive from the chain, signed so they cannot be edited after the fact, on a public page that is honest about which part of the backing is on-chain and which part rests on an attestor. That layer is what the GENIUS-readiness kit is, and it is priced for the issuer the Big Four will never call back.
What a miss would mean
Suppose the agencies do not finish by July 18. It is a real possibility: as the run-up begins the Federal Reserve has not proposed, and no final rule exists. The honest answer is anticlimactic, and that is itself the point. The one-year deadline in Section 13(a) is directory. The statute attaches no penalty to a miss, names no rule that springs into force automatically, and sets no interim standard. Missing it does not trigger anything; it simply means Section 20's 120-day clock never starts, the January 18, 2027 backstop becomes the operative date, and until then the state of United States reserve disclosure is the pre-GENIUS status quo: the better issuers publish monthly attestations because they choose to, the rest publish what they like, and nothing federal yet separates the two. If you have been reading "GENIUS passed" as "stablecoin reserves are now federally policed," the run-up is the moment to notice those are different sentences, and will stay different sentences on either side of July 18.
Where Kerne sits, and what the run-up is actually for
We say where we stand so none of this floats. kUSD is not a payment stablecoin and we do not present it as one: it is a strategy-backed synthetic dollar, which by the same reading that lets this category pay yield puts it outside the Act's payment-stablecoin box, and therefore outside its floor. So we do not reach for a regulatory label we would not qualify for. We build the checkable layer instead. kUSD's reserve leg lives on Base and is readable with raw on-chain calls, and we publish an hourly Proof of Reserves signed with an EIP-191 key that you re-derive yourself against the chain, with no attestor in the path, at /verify and /api/por. The honest boundary is the one we disclose everywhere: the delta-neutral hedge runs on Hyperliquid, a single venue, and that leg is self-reported and signature-bound rather than independently re-derivable, an independent attestation of it is being scoped, we are pre-audit and at Genesis scale, and the skUSD yield is a live model rather than a large realized distribution. Our open gaps are listed rather than hidden at /legible, and our place in the field, including where it does not lead, is in the synthetic-dollar scorecard.
If you issue a dollar this rulebook will reach
The examination is not the only thing your holders will look at, and it is the one thing you cannot build yourself. The GENIUS-readiness kit stands up the part you can: a hosted verify page for your token, scheduled machine-signed reads of your disclosed reserve addresses, and an embeddable freshness badge for your own site, live within five business days, $499 setup then $99 a month. It is not the statutory examination and it does not make you compliant with anything; it is the layer a holder can check, which in a market this crowded is what tells two GENIUS-ready pages apart.
The rules will land, or they will slip, and we will write the rule-day analysis the day they actually arrive rather than the day the calendar says they should. Until then the useful thing to do with the run-up is the same on either branch: read Section 4, check whether the dollar in front of you even falls under it, and for the part no rule covers, confirm you can verify the backing yourself. That is the check the statute does not run for you.
Figures and status are as of July 6, 2026, and nothing here is legal or investment advice. The GENIUS Act was signed July 18, 2025 as Public Law 119-27; the one-year rulemaking mandate in Section 13(a), the four primary Federal payment stablecoin regulators and the Treasury Secretary's distinct anti-money-laundering and sanctions mandate, the interest prohibition in Section 4(a)(11), the Section 4 reserve, disclosure, monthly-examination, certification, redemption, and Bank Secrecy Act requirements, and the Section 20 effective date (the earlier of January 18, 2027 or 120 days after final rules) are per the enacted text on Congress.gov and the summary by Morgan Lewis. The rulemaking timeline (the OCC proposed rule published March 2, 2026 with comments closed May 1; the joint FinCEN and OFAC and the FDIC proposals published April 10, 2026 with comments closed June 9; the Federal Reserve not having published its own proposal as of late June; and no final implementing rule as of publication) is per the Chapman and Cutler GENIUS Act rulemaking tracker, the Federal Register, and the agencies, re-verified on the publication date. The reading that derivatives-backed synthetic dollars fall outside the payment-stablecoin definition is the prevailing practitioner interpretation, not settled law. The Forbes reference is Zennon Kapron, "Ethena's USDe Pays Yield Legally, And The GENIUS Act Has No Answer For It," Forbes (June 15, 2026), a contributor column. Falcon Finance and Anchorage Digital's fUSD, marketed as GENIUS-ready with a monthly Deloitte attestation and launched in late May 2026, and Falcon's separate USDf synthetic dollar at roughly 1.26 billion dollars outstanding per DefiLlama, are per Falcon's and Anchorage's own announcements and public trackers; Kerne is not affiliated with any company named. Kerne's own claims resolve to live endpoints: the hourly signed Proof of Reserves at /api/por/signed, its on-chain leg at /api/por, and the live risk surface at /api/risk-status. A /verify pass proves an attestation is authentic and fresh; it is not an audit and not a solvency opinion.
Verify it yourself
Run the same check on any reserve, or have it run for you.
Paste any issuer's signed attestation into the free verify tool and recover the signer, rehash the figures, and check freshness in your own browser. For a machine-signed, point-in-time read of an address you name, delivered on the page in about two minutes, the instant self-serve read is $29; a human-reviewed read is $149. A teardown like this one, commissioned on any target you name, is $499. An independent read of a counterparty you hold or allocate to is $2,500. Attestation tooling, not an audit, and not a solvency opinion.