The GENIUS Act Rules Just Missed Their Own Deadline. The Federal Floor Is Not Under You Yet. You Can Still Verify Anyway.

On July 18, 2026, a statutory deadline passed quietly. Section 13 of the GENIUS Act, the first federal payment-stablecoin law, directed each primary federal regulator and the Treasury to promulgate implementing regulations, in the statute's words, not later than one year after the date of enactment. The Act was signed July 18, 2025. The year is up and, as of publication, there is not one final implementing rule from any of those agencies in the Federal Register.

This is the companion piece we hoped not to publish. We wrote a rule-day analysis for the branch of history where the rules landed on time, and this one for the branch where they did not. Since you are reading this one, the honest first sentence is this: the federal floor under stablecoin reserve transparency, the thing the last year of "regulated dollars are coming" coverage has been pricing in, is still a promise, not a rule in force.

What actually happened, without the drama

The miss was visible months out, and it is worth being precise rather than breathless. By deadline week every agency had proposals on the record and none had a final: the OCC's core proposal went to the Federal Register in early March with comments closed May 1; the FDIC's followed in April; Treasury, FinCEN and OFAC ran a joint anti-money-laundering proposal with comments closed in June; the NCUA's implementation proposal took comments through July 17, one day before the statutory deadline. The Federal Reserve joined a multi-agency customer-identification proposal in June, whose comment window runs to August 21, but as of this writing it has not proposed its own core GENIUS framework rules at all. You cannot finalize what you have not proposed, and you generally do not finalize while a comment window is still open. The arithmetic of a miss was locked in before the date arrived.

Two more facts keep the temperature down. The statute attaches no penalty and no fallback to the deadline; Section 13 simply says shall, and nothing happens juridically when shall meets a slipped calendar. And agencies miss statutory rulemaking deadlines routinely, across administrations and subject matters; this one is notable because of what the rules are for, not because lateness is rare.

What the miss actually changes for a holder

The GENIUS Act's substantive core, for anyone holding a dollar token, is a disclosure regime: a permitted issuer must publish the composition of its reserves on its website every month, have each month-end report examined by a registered public accounting firm, and have its CEO and CFO certify the report to the regulator, with criminal liability for a knowingly false certification. We walked that floor, and why a floor is not a proof, in the piece written for the other branch of this fork. The mechanics live in the statute itself, so they are coming regardless.

But they are not here. The Act takes effect on the earlier of January 18, 2027 or 120 days after the primary regulators issue final regulations. No final regulations means the 120-day clock has not started, and every week of slippage pushes the operative date toward the January 2027 backstop, arithmetic the payments press ran before the date even arrived. Until then, the roughly $300 billion stablecoin market (about $304 billion by CoinGecko's count on deadline day) runs on exactly the status quo the Act was written to fix: the better issuers publish monthly attestations voluntarily, the rest publish what they choose, and nothing legally separates the two. If you have been treating "GENIUS passed" as "stablecoin reserves are now federally policed," today is the day to notice those are different sentences.

There is a second-order effect worth naming plainly. The yield-bearing synthetic dollars we cover in this series mostly sit outside the payment-stablecoin definition anyway, by the prevailing legal reading, so the floor was never coming for them. The miss does not change their position. It changes yours: the regulated tier you might have rotated toward, on the theory that it now carries a federally examined floor, does not carry it yet. The market did not wait for the rulebook either. As perpetual funding compressed, the largest dollar in that uncovered category, Ethena's USDe, spent this year shifting the bulk of its backing from the crypto basis trade toward lending and real-world assets, and its staked yield fell to about the level of Treasuries; Forbes noted in June that the Act has no answer for the yield a token like it pays. The rulebook and the market are moving on different clocks, and neither settles what you can check yourself.

The check that does not wait for a rulemaking calendar

Here is the useful thing about holder-side verification: it has no effective date. A reserve you can re-derive from the chain yourself does not care whether the OCC's final rule is out. That is the axis we keep returning to, attestor-required versus attestor-optional. A monthly examined attestation, when it eventually binds, is a professional proving the number at a point in time while you read their result. A reserve readable from raw on-chain data, checked against a signed proof, is a number you re-derive whenever you want. The first is a floor that is currently not in force. The second is available this afternoon.

The first half of this year is that argument in numbers. Forbes, reading CertiK's half-year data, put H1 2026 losses at $1.32 billion across 344 incidents, and the composition matters more than the total: contract-code bugs accounted for about $151.6 million of it, while compromised wallets and keys were the costliest vector at $444.5 million, and the two largest single losses, Kelp DAO at $291 million and Drift at $285 million, were operational and infrastructure failures rather than code vulnerabilities. A monthly attestation is not built to surface an operator-side failure inside the month it happens. A signed, frequently refreshed read of the reserves, checkable by holders, is the layer that narrows that window, and the signing itself has to be treated as attack surface: our own signed Proof of Reserves shipped a verification defect in May, which we found, fixed, and wrote up in public.

We say where we stand so this is not an abstraction. kUSD, our synthetic dollar, is not a payment stablecoin and we do not present it as one; the missed deadline neither helps nor hurts its regulatory position, and nothing here is an offer of it to anyone. The relevant part is the method: kUSD's collateral 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 at /verify, with no attestor between you and the on-chain figure. The honest boundary is disclosed in the same breath: the delta-neutral hedge leg runs on Hyperliquid, a single venue, self-reported and signature-bound rather than independently re-derivable, an independent attestation of that leg is being scoped, and we are pre-audit at Genesis scale with a live modeled yield, in the low teens while funding is positive and normalizing toward the high single digits through a cycle. The point is not that kUSD escapes trust. It is that the part of it you can check does not depend on any agency's calendar, and the part you cannot check is named on the same page.

If you want the same question answered about anything you hold, the free tool at /verify-anything runs the on-chain half on any token, the synthetic-dollar scorecard ranks the field on exactly this axis, including where kUSD falls short, and our own open gaps are at /legible. And because yield claims decay even faster than reserve claims, we publish the honesty index: a signed, hourly board of advertised versus realized yield across the field, updated on a clock no regulator sets, with our own numbers listed first and, at the moment, worst.

What to watch next

The rules will land; the only question this post answers is that they did not land on time. When they do, the 120-day clock starts, the floor becomes real, and we will publish the rule-day analysis waiting in the other branch of this fork. Watch the Federal Register, not the press releases, and when the monthly examined disclosures start appearing, read them. Then do the thing the examination cannot do for you: check whatever can be checked without anyone's permission, and know exactly where the checkable part ends. That line, not the rulemaking calendar, is what protected holders through every failure this series has covered.

Status is as of the publication date and nothing here is legal or investment advice. The GENIUS Act was signed July 18, 2025 (Public Law 119-27); Section 13's one-year rulemaking mandate, Section 20's effective date (the earlier of January 18, 2027 or 120 days after final regulations), and Section 4's monthly disclosure, monthly examination by a registered public accounting firm, and CEO and CFO certification with criminal penalties by reference to 18 U.S.C. 1350(c) are per the enacted text on govinfo.gov. Rulemaking status is per the Chapman and Cutler GENIUS Act rulemaking tracker and the Federal Register, re-verified on the publication date; the stablecoin market figure is the CoinGecko stablecoins-category capitalization read on July 18, 2026 (cross-checked against DefiLlama and RWA.xyz, which read between $299 billion and $308 billion the same day); H1 2026 loss figures are per CertiK's Hack3d H1 2026 report as reported by Forbes on July 17, 2026; the reading that derivatives-backed synthetic dollars fall outside the payment-stablecoin definition is the prevailing practitioner interpretation, not settled law. Kerne's 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.