Phantom and Hyperliquid take their case to the CFTC: carve out the code, not the company

On 9 July 2026, Phantom, the consumer wallet best known for the Solana logo that lives in millions of browser extensions, and Hyperliquid, the onchain perpetual-futures venue whose order book has drawn more flow than some central limit order books, filed a joint petition asking the Commodity Futures Trading Commission to redraw the line between a derivatives intermediary and a piece of code. The two firms want the agency to confirm that blockchain developers and non-custodial wallet providers do not, by virtue of writing or distributing software, become the kind of futures commission merchants, introducing brokers or swap dealers the Commodity Exchange Act was built to police.

The filing lands at a moment when the CFTC is short on staff, short on cases closed under its existing digital-asset remit, and long on questions it has so far preferred to answer through enforcement actions rather than rulemaking. Phantom and Hyperliquid are betting that the agency's 2026 priorities, which the new chair has repeatedly described as modernising the rule book for non-intermediated markets, are less a talking point than a working instruction.

What the petition actually asks

The substance is narrow and technical, and that is the point. According to Cointelegraph's 9 July 2026 write-up of the filing, the companies urged the regulator to exempt blockchain developers and non-custodial wallet providers from rules designed for traditional financial intermediaries, on the grounds that publishing open-source code or running a self-custody interface does not make a software publisher the principal, counterparty or custodian of anyone else's trade. Hyperliquid's policy arm has separately framed the request as a clarification: in their reading, existing CFTC interpretive letters already point this way, but the agency has not yet committed that reading to a binding order, and the gap is the gap that scares institutional underwriters away from routing customer flow through onchain venues.

For Phantom, the calculus is simpler and more existential. A consumer wallet lives one regulator's interpretation away from being treated as a money services business, a money transmitter, a broker, and now, potentially, a derivatives intermediary. Each label pulls in different reporting, capital, cybersecurity and supervision obligations, and each one was drafted assuming the regulated entity holds client funds, signs client agreements, or sets the price of the trade. Phantom does none of those things. Its code signs transactions on behalf of the user, and the user, not the wallet, controls the keys. Hyperliquid runs a matching engine and a vault, and is more comfortable in the intermediary frame, but its policy arm is pushing for the same boundary in the same petition because the venue it operates cannot fully decouple its product roadmap from the regulatory perimeter applied to the software clients that point at it.

The agency they are petitioning

The CFTC is, as of mid-2026, in the unusual position of being asked to write rules for a market that has, in some segments, already eclipsed the regulated futures exchanges it is supposed to oversee. Perpetual futures on offshore venues have, on multiple days in the past year, settled more notional volume than the CME's bitcoin and ether futures contracts combined. The agency can respond by treating that volume as a compliance problem, which is what its enforcement division has largely done since 2023, or by treating it as a jurisdiction problem, which is what a petition like this one implicitly does: if the venue, the matching engine, and the user interface are each separately characterised, then the agency has to decide which one of them it actually regulates, and on what statutory hook.

The chair's public comments over the past quarter have leaned toward the second framing. The agency's 2026 budget request asks for additional headcount specifically to modernise the rule book for non-intermediated markets, and the most recent enforcement-closure statistics show a higher share of cases resolved with monetary penalties and remediation rather than the precedent-setting litigation the previous commission pursued. That is the operating environment into which Phantom and Hyperliquid have chosen to drop a petition that is, in effect, an invitation: please write the rule you have been talking about writing.

The structural argument underneath

Beneath the case-specific drafting is a larger question about where the perimeter of US derivatives law sits in a world where the matching engine is open source, the custody layer is the user's own hardware wallet, and the front end is a website that can be hosted anywhere. The traditional answer is that the regulator regulates the entity it can reach, and the entity it can reach is usually the one with a US bank account, a US server, and a US lawyer. The Phantom-Hyperliquid filing asks the agency to take a different cut: regulate the function the entity performs, not the entity itself, and exempt functions that no human or firm performs because the software performs them.

That move, if granted, would do more than resolve the immediate question of whether a self-custody wallet is a derivatives intermediary. It would set a precedent for how the agency characterises decentralised infrastructure across spot crypto, tokenised securities, and the onchain versions of interest-rate swaps and credit derivatives that are now being prototyped by the same set of firms that built the perpetual-futures venues. It would also draw a line that the Securities and Exchange Commission, which has so far preferred to regulate through enforcement under existing Howey-framework tests, will eventually be asked to either match or contest. Two regulators writing two different definitions of the same piece of code is the kind of jurisdictional arbitrage opportunity the industry has spent the last decade learning to exploit.

What the counter-narrative sounds like

The counter-narrative is straightforward and is not fringe. Consumer-protection groups and a meaningful slice of the state regulator community argue that a wallet which recommends tokens, ranks decentralised applications, and routes swaps through an aggregator is performing the function of a broker even if the keys never leave the user's device. Under that reading, the petition is less a clarification than an attempt to lock in a definition that benefits infrastructure providers while leaving the retail user, who does not read CFTC petitions, with the same set of counterparty and operational risks they already face. The trade press covering this beat, including Cointelegraph's own 9 July piece, has flagged the same tension: any rule that exempts wallet developers has to specify, in detail, what a wallet can and cannot do without losing that exemption, and the line between routing a user's swap and recommending a counterparty is doing a lot of work.

What to watch next

The CFTC has, in past cycles, taken between four and fourteen months to act on substantive petitions of this kind, and the most consequential step is usually the Federal Register notice seeking public comment. The next concrete data points are whether the agency dockets the filing as a formal rulemaking proceeding, whether it circulates an advance notice of proposed rulemaking ahead of the formal comment window, and whether the SEC, the Treasury and the prudential bank regulators file parallel comments that signal either alignment or jurisdictional friction. Phantom and Hyperliquid have, in effect, put a marker down in front of all three.

The honest read is that the petition is more likely to produce a partial answer than a clean one. The most probable outcome is a definitional carve-out for non-custodial software that comes with conditions: no custody, no discretionary routing, no fee revenue from the trade itself, and a series of disclosures about counterparty risk that look uncomfortably close to the disclosures a broker would have to make. That is not what the petitioners asked for, and it is not what their opponents want either, which is usually the shape a regulatory compromise takes in a market that has outgrown the regulator's vocabulary.

This publication treats onchain-derivatives rulemaking as a structural question about where the US regulatory perimeter ends in markets that increasingly organise themselves around software rather than firms. Where wire coverage has focused on the personalities behind Phantom and Hyperliquid, Monexus focused on the carve-out itself, on what it would and would not cover, and on what the SEC's parallel silence on the same boundary implies about the next twelve months of US crypto rulemaking.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://t.me/CryptoBriefing