On August 21, 2026, the United States Securities and Exchange Commission (SEC) published a 400-page proposed rule called Regulation Crypto Assets. Until now, the SEC has only ever fitted crypto tokens into rules built for something else – a 1946 court test meant for ordinary investment schemes, and exemptions written for company shares or real estate trusts. This is the first exemption written specifically for crypto tokens. It sets out how a project can raise money from the public without a full securities registration, and what an investor must be told before putting money in.
This matters outside the US as well. A large share of global token launches, wherever they happen, still keep US securities law in view, because so much capital, so many exchanges and so much developer activity is connected to the US market in some way. When the SEC changes how token fundraising works at home, it changes the reference point that other markets and other regulators tend to watch and compare against. This piece goes through what the proposal actually says, why the SEC felt the need to write it, and what it could mean once it takes final shape.
What the SEC Could Not Answer
The SEC has never lacked authority over crypto. What it lacked was an answer to one question: at what point does a token stop being a security? Without that answer, a project could not tell whether raising money required full registration, and many either avoided US investors or set up offshore.
The difficulty traces to a 1946 Supreme Court case, SEC v. Howey, which treats an arrangement as a security if a person puts in money as part of a common venture and expects profit mainly from the effort of others. That works where status is fixed. A token is different. When first sold, it usually comes with a promise from the issuer to build the network, manage the treasury and keep developing it, and that promise is what makes the arrangement look like an investment contract. Once the network runs on its own, without the founding team, the promise no longer exists. A company share does not behave this way. A token can begin inside the definition and later fall outside it, with nothing about the token itself having changed.
The clearest proof of the mismatch is in the SEC’s own litigation record. In the Ripple case, a federal court held that XRP sold directly to institutions was a security, while the same token sold on exchanges was not. One asset, two answers, decided years after the sales happened. Industry participants called the old approach regulation by enforcement, where rules only became clear once someone was sued. What is easy to miss is that the reverse criticism also exists: SEC Commissioner Caroline Crenshaw described the agency’s newer habit of dropping crypto cases while promising future rules as regulation by non-enforcement. Both criticisms point at the same defect. The SEC was making policy through individual decisions rather than published rules, and it could do that in either direction.
Two Routes to Raise Money, and Disclosure Built for Tokens
The answer is a pair of new exemptions, but the money limits are the less interesting half. The more revealing change is that the SEC has written disclosure requirements aimed at what a token actually is, rather than making issuers fill in forms designed for a company selling shares in a factory.
Both routes cover what the SEC calls a covered investment contract – a token sold along with a promise from the issuer to keep building the project, which describes most token launches. The startup exemption allows up to $5 million over four years, aimed at early-stage teams who might otherwise run a pre-sale in a legal grey area or set up outside the US. The fundraising exemption, modelled on Regulation A, allows up to $75 million in any 12-month period; past a certain size, issuers must file audited or reviewed financial statements and periodic reports with the SEC, scaled down for a token rather than a share. Both routes require plain-language disclosure on the team and the risks, filed on EDGAR, the SEC’s public filing database, where any investor can read them. Using either route also does not protect an issuer who lies. If a project makes false or misleading statements while raising money, investors and the SEC can still sue it for fraud exactly as before; the exemption removes the registration requirement, not the duty to tell the truth.
The disclosure content follows the same logic: what is being built, how supply and allocation work, who controls the code, what rights holders have, and what could go wrong. This is a deliberate break from Regulation A and Regulation D, the frameworks these exemptions borrow from, both written for shares and bonds where ownership, control and profit rights are fixed on the day of sale. A token’s rights shift as a network decentralises, which is exactly the gap Ripple exposed. Building disclosure around that movement is the SEC closing the uncertainty its own litigation could not resolve.
Who Decides What Counts as Important
Here the first handover appears. The rule avoids a checklist and instead names ten broad topics an issuer must cover, including the token, the team and its conflicts of interest, the network’s development plan, the source code and security, supply allocation, governance rights and risk factors. Within each topic, the issuer decides what is material, meaning important enough that an investor would want to know it.
Leaving that call to the issuer creates a predictable gap. With no minimum content, two projects raising identical amounts can file disclosure of very different depth and both satisfy the rule. A weak team has every reason to describe its own risks in the thinnest terms the topic list allows, and an investor cannot tell thin disclosure from honest disclosure until something fails. This is the same weakness the SEC set out to escape, moved from the classification stage to the disclosure stage. Its answer is a disqualification rule barring issuers with a record of fraud, certain convictions, or prior SEC action from either exemption. That screens out known bad actors and does nothing about a first-time issuer who simply says as little as permitted, which is the more likely failure in practice.
How a Token Stops Being a Security
On the central question, the proposal delivers a defined exit. The safe harbor applies once a token no longer depends on the founding team, meaning the network runs, updates and secures itself without any single party’s continued effort. Two conditions apply: the issuer must have finished or permanently stopped the managerial work it promised at launch, and it must file a transition report, Form TR, announcing this. Meet both, and the token is treated as free of the investment contract, able to trade the way Bitcoin already does.
The trouble is that the test is hard to pass and nobody checks the answer. Legal analyses of the text have flagged that essential managerial efforts is fact-heavy: detailed public statements about development milestones, funding or timelines will likely count as ongoing managerial effort, which makes the exit a pathway rather than a switch. So an issuer must judge a genuinely difficult legal question about itself, and Form TR is its own declaration, not something the SEC verifies before the token trades freely. A team could file while still quietly steering the code or the treasury. The combination is awkward: a hard test with no examiner tends to produce confident filings from those who should be cautious and caution from those who would have qualified.
One Regulator Instead of Fifty-One
The provision that saves issuers the most money is also the most likely to be fought. Each US state runs its own securities law, known as Blue Sky law, with separate registration and review, so a project raising nationally has had to satisfy the SEC and then up to fifty more desks. Under the new rule, anyone buying a covered investment contract counts as a qualified purchaser, and federal law already bars states from imposing registration requirements on sales to that group. One regulator replaces fifty-one, with the protection extending to later buyers as long as the issuer keeps filing.
This has been fought before. When the SEC used the same mechanism for Regulation A+ in 2015, state regulators objected and Montana and Massachusetts sued. The states lost at the DC Circuit in 2016, but their argument was never answered: state regulators sit closer to local investors, act on complaints faster, and can freeze an offering within days where federal action takes far longer. Removing their registration review does not remove the fraud they were catching; it removes a filter that caught it early. For crypto, where an offering reaches every state within hours of going live, that timing gap matters more than it did for a conventional small-company raise. Raising the ceiling to $75 million at the same time means federal disclosure now carries alone the protective load two layers used to share.
A Sweeping Rule That Nobody Inside the SEC Argued With
A less discussed aspect of this proposal is the circumstances in which it was approved. It advanced by written vote from Chairman Paul Atkins and Commissioners Peirce and Uyeda without dissent, at a point when the Commission had no member likely to take an opposing view. Caroline Crenshaw, its last Democratic commissioner and a consistent sceptic on crypto, left in January 2026. For decades the SEC has kept a minority voice on the Commission, not because the law requires it, but because a commissioner writing in opposition prompts the majority to set out its reasoning on the record before a rule is published.
The effect of that shows up in the text itself. Each of the three handovers described above – materiality decided by the issuer, security status self-certified, state review removed – turns on the same unanswered question: who checks that any of it was done properly. A written objection would ordinarily have forced that question into the open before publication, and the absence of one is the simplest explanation for how broadly the framework is drawn. It also means the scrutiny has not disappeared, only moved outside the Commission, to state regulators, comment letters and whichever Commission inherits the rule. Chair Atkins has said he hopes to future-proof these changes so the pendulum does not swing back, but the opposite reading seems more likely: a rule carries less weight against a future reversal when the record holds no tested defence of its weakest parts. The economic analysis leaves a related gap. Compliance costs do not fall with the size of a raise, so a team raising $3 million carries close to the same fixed burden as one raising $50 million – which points to a framework most usable by the issuers who needed an exemption least.
How Japan and the EU Compare
Three large jurisdictions moved on crypto this year, each picking a different point between easier access and tighter oversight. Japan moved crypto into its main financial instruments law, raising standards for exchanges while lowering the tax rate as an incentive, and kept exchanges under continuous supervision rather than one-time approval. The EU’s MiCA became fully enforceable across all 27 member states from July 1, 2026, replacing national regimes with a single licence that passports across the bloc. The US removes a layer of review the way MiCA replaced twenty-seven national ones, but through preemption rather than agreement, leaving the SEC as sole reviewer. The EU shows the cost of that design: French, Austrian and Italian regulators have accused firms of picking the member state with the lightest review and then operating everywhere on that approval. A single approval route helps compliant issuers and is harder to police for everyone else, which is the same risk inside the move from fifty state reviewers to one.
What Happens Next
Comments close on October 20, 2026, and the SEC has invited views on whether $5 million is the right startup limit and whether the safe harbor conditions are workable. On the Regulation A+ precedent, state regulators are the likeliest organised opposition, and the $75 million ceiling paired with full preemption is the provision most exposed to revision. The proposal also leaves open how these offerings interact with exchange, broker and dealer registration, which it does not resolve. That gap was meant to be filled by the CLARITY Act, which failed a Senate procedural vote on September 15, so it now stays open.
The direction, though, is set. After years of policy made through enforcement decisions in one direction and dropped cases in the other, the SEC has put its reasoning in a public document with numbers and procedures attached. Its weaknesses are real, and they share one shape: at each difficult point the rule trusts the issuer and hopes the market or a lawsuit catches the rest. But they are weaknesses of a written rule that can be argued with and amended, which neither of the previous approaches offered. For founders, for investors, and for regulators elsewhere including in India watching how the largest capital market handles this, the shift from case-by-case judgement to a published rulebook is the part worth following – and the comment file, not the proposal, is where it will be decided.
References
- U.S. Securities and Exchange Commission, Regulation Crypto Assets – Proposed Rule – https://www.sec.gov/files/rules/proposed/2026/33-11434.pdf
- SEC, Application of the Federal Securities Laws to Certain Types of Crypto Assets and Certain Transactions Involving Crypto Assets, Release No. 33-11412 (March 17, 2026)
- President’s Working Group on Digital Asset Markets, Strengthening American Leadership in Digital Financial Technology (July 30, 2025)