SEC Innovation Exemption and Token Safe Harbor: A Founder's Guide to Regulation Crypto Assets (Updated August 2026)

By Chanté Eliaszadeh | May 10, 2026 | Updated August 19, 2026

What Is the SEC Innovation Exemption? The Three-Part Framework at a Glance

For the past decade, Howey was the test the SEC used to say no. On March 17, 2026, Chair Paul Atkins announced an Innovation Exemption that uses Howey to say yes—three new pathways with the doctrine’s own fourth prong, the cessation of essential managerial efforts, as the registration architecture’s load-bearing element.¹ The rulemaking package went to OIRA pre-publication review on April 6, 2026,² blew through the agenda’s own April target month, drew a formally noticed Commission vote on August 14, 2026 that produced no release, and then landed on August 18, 2026 as a 402-page proposing release, Release Nos. 33-11434 and 34-106150.³ The dated status analysis below tracks exactly where it now sits.

The framework has three parts: a Startup Exemption for early-stage raises Atkins described on the order of $5 million over roughly four years, a Fundraising Exemption for mid-scale raises up to $75 million per twelve-month period, and an Investment Contract Safe Harbor that lets tokens transition from security to non-security status when the issuer’s essential managerial efforts cease. Both dollar figures survived into the proposal as rule text, and the fundraising exemption arrived split into two tiers, $20 million and $75 million, that the March remarks did not describe. For crypto founders who spent the last decade in regulatory limbo, this is what they asked for. None of the three pillars is final, and structural challenges to the safe harbor’s legal authority are already being prepared—founders who treat the rule as litigation-proof will be the ones surprised.

The taxonomy the SEC and CFTC adopted the same day Atkins delivered his Token Safe Harbor speech answered the question that has paralyzed crypto founders for years: what is my token?⁴ The Innovation Exemption answers a different question: how do I bring it to market? For founders sitting on capital decisions today, the answer reshapes Reg D, Reg S, and Reg A+ economics for the next generation of token offerings.

This article is the founder-grade decision tree for the proposal—less another institutional explainer⁵ than a guide to which pathway fits which token archetype, what each exemption requires, and where the rule’s legal vulnerabilities sit.

Key Takeaways

• The proposal is out; the comment clock is not running yet. The Commission issued Regulation Crypto Assets on August 18, 2026 (Release Nos. 33-11434 and 34-106150, File No. S7-2026-27). Comments are due sixty days after Federal Register publication, which had not occurred as of August 19.³
• “Innovation exemption” now names two different SEC workstreams. The one in the August headlines is a Trading and Markets rule about trading tokenized securities. This guide covers the Corporation Finance offering regime, RIN 3235-AN38.
• Blue-sky preemption is in, through section 18(b)(3). The proposal defines “qualified purchaser” so that state registration and qualification are preempted for offerings under the regulation and for certain secondary transactions—the most consequential question this guide has flagged since May, now answered.³
• The thresholds are rule text now, and there are more of them. $5 million over four years for the startup exemption; the fundraising exemption splits into a $20 million Tier 1 and a $75 million Tier 2 per 12-month period, with affiliate-resale sublimits and no financial-statement assurance requirement at Tier 1.³
• The release recites its authority but never argues it for the safe harbor. The agenda’s “Not Yet Determined” entry is superseded by an omnibus authority recital, yet the safe-harbor discussion runs on purpose—clarity, certainty, capital formation—and not on power. That gap is the APA challenge this guide analyzes.³
• Raise on the live stack now. A proposal confers no relief. Reg D 506(b)/(c), Reg S, and Reg A+ Tier 2 work today; build election optionality in and treat adoption as upside, never as the plan.

Which SEC Innovation Exemption Are You Looking For? Two Different Rules, Same Name

There are two, they come from different divisions of the SEC, and they do different things. This guide covers the offering one.

The name founders arrive with—“the SEC innovation exemption”—was first used by Chair Paul Atkins at the Commission’s DeFi Roundtable on June 9, 2025, and in that usage it means a Division of Trading and Markets workstream on limited on-chain trading of tokenized securities. Commissioner Mark Uyeda described it that way at SEC Speaks on March 19, 2026: “We are developing an innovation exemption that would facilitate limited trading of certain tokenized securities.” Jamie Selway, Director of the Division of Trading and Markets, said on April 13, 2026 that his division “is currently working on an ‘innovation exemption’ recommendation to the Commission to allow certain trading venues to trade tokenized securities.” That is the workstream that carried mainstream financial headlines in August 2026.

This guide is about the other one: the token offering framework Atkins outlined on March 17, 2026, which the SEC’s own rulemaking agenda calls Crypto Assets (RIN 3235-AN38) out of the Division of Corporation Finance, and which the Commission’s August 2026 meeting notice describes as “a tailored offering regime for certain investment contracts involving crypto assets.” It governs how a token is sold, not where it trades afterward.

If you arrived from a headline about tokenized stocks or a delayed tokenization vote, you want the trading exemption, not this page. Both are unfinished, and neither is law.

Is the SEC Token Safe Harbor in Effect Yet? (Status as of August 19, 2026)

Not yet—but the wait for rule text is over. On August 18, 2026 the Commission proposed Regulation Crypto Assets, Release Nos. 33-11434 and 34-106150, File No. S7-2026-27, RIN 3235-AN38: a 402-page proposing release that would create a new 17 C.F.R. part 228 and put all three pathways into text a founder can read.³

A proposing release is not a rule. Nothing in it may be relied on, nothing may be offered or sold under it, and every figure in it can change before adoption. What changed on August 18 is the quality of the information, not the state of the law—founders now reason from rule text instead of from a speech.

The regulation is organized in five subparts, and the shape is close to what this guide predicted from the March remarks:

The comment clock has not started. The Commission issued the release on its own site on August 18; the release publishes in the Federal Register separately, and comments are due sixty days after that publication date—the release carries a bracketed placeholder where the date will go. As of August 19, 2026, no Federal Register document for File No. S7-2026-27 had appeared.⁶ Track the Federal Register document, not the press release; the sixty days run from the former.

The enforcement backdrop that made founders keep asking is unchanged. SEC crypto enforcement fell from 33 new actions in 2024 to 13 in 2025, and zero of the eight actions initiated under Chair Atkins alleged registration violations—every one alleged fraud (Astraea Counsel analysis of the firm’s Crypto Enforcement Tracker dataset of SEC and CFTC actions, 2024-2026). The agency stopped suing on registration theories and started writing rules. The rules have now arrived in draft. For the political arc behind that shift, see what the SEC’s new posture means for startups.

A proposal is a document, not a permission. Until a final rule takes effect, the live exemption stack is still the only thing a founder can actually raise under.

What Three Crypto Rules Are on the SEC’s 2026 Rulemaking Agenda?

Three separate rulemakings: RIN 3235-AN38 (crypto-asset offering exemptions—the token safe harbor package this guide covers), RIN 3235-AN48 (broker-dealer capital and custody for digital assets), and RIN 3235-AN49 (crypto market structure).⁷

Trade coverage collapses all three into “the SEC’s crypto rule”; they are distinct proposals with distinct timetables. The Spring 2025 agenda edition carried an April 2026 NPRM target for the offering-exemption rule—a target that passed—and the current 2026 edition, published July 7, 2026, carried a July 2026 target for all three; the offering rule arrived on August 18, and the other two have not.⁷

Only one of the three has left the target column. The other two remain what every “Proposed Rule Stage” cell means: a date the agency wrote down, not a rule.

What Atkins Actually Proposed: A Three-Part Framework

The framework is not one rule. It is three. Each piece addresses a different scale of token offering, carries different requirements, and interacts differently with the existing exemption stack. Founders need all three, because the right pathway depends on offering size, network maturity, and how aggressively the project pushes toward decentralization.

Before the mechanics, a status-precision device the trade coverage rarely runs: where each headline number actually lives. Until August 18 every figure in this framework lived in a speech. They now live in proposed rule text, which is a real upgrade in reliability and still not law—a proposed figure can move between proposal and adoption, and several of these are the subject of express requests for comment.

Every threshold in this article is now sourced to the August 18, 2026 proposing release rather than to the March remarks—and every one of them can still change before adoption.

The Startup Exemption: $5 Million Over Four Years

The startup exemption is the de minimis pathway for early-stage projects. Proposed Rule 200 would exempt from Securities Act section 5 offerings of up to $5 million of covered investment contracts during a four-year period, subject to principles-based narrative disclosure made available to investors.³ The framing is founder protection: a project at seed-and-Series-A scale should not have to retain specialist securities counsel and run a full Reg D 506(c) verification process to launch a token.

The $5 million cap is small relative to Reg A+ Tier 2’s $75 million ceiling, 17 C.F.R. § 230.251 et seq., but the four-year duration window distinguishes it from instantaneous registration exemptions: it functions as a runway during which the issuer can develop the network, distribute the token, and—if the project succeeds—graduate to the fundraising exemption, a registered offering, or the safe harbor’s status-based exit. Disclosure is principles-based—narrative, whitepaper-style materials covering the project, team, token economics, and risks—rather than prescriptive Form 1-A or Form C requirements. Antifraud liability would remain fully preserved under Section 17(a) of the Securities Act, 15 U.S.C. § 77q, and Rule 10b-5; AML and KYC would likely be integrated through the Bank Secrecy Act framework, 31 U.S.C. § 5311 et seq., cross-referencing the GENIUS Act’s Permitted Payment Stablecoin Issuer obligations.

The most consequential question this guide flagged before the proposal—whether the exemption would carry covered-security treatment under Section 18(b)(3) or 18(b)(4) of the Securities Act, 15 U.S.C. § 77r—is answered. The Commission chose (b)(3), the qualified-purchaser pathway Reg A+ Tier 2 uses, and proposes to define “qualified purchaser” so that state registration and qualification are preempted both for offerings under the regulation and for certain secondary transactions in the same asset.³ That is the difference between fifty-state blue-sky review and single-track federal preemption, and it is the single most valuable thing in the proposal for a founder planning national distribution. Read the conditions before relying on the headline: preemption follows the exemption, so an offering that falls out of the exemption falls out of preemption with it.

For founders raising under $5 million, the startup exemption is the cheapest pathway on the board as proposed—preemption, principles-based disclosure, and a four-year runway. It is also not yet available, and every one of those terms is open for comment.

The Fundraising Exemption: $75 Million Per Twelve Months

The fundraising exemption is the mid-scale pathway, and the proposal splits it in two. Proposed Rule 300(a) sets a Tier 1 of up to $20 million per 12-month period, of which no more than $6 million may be offered by affiliate selling securityholders, and a Tier 2 of up to $75 million, of which no more than $22.5 million may be affiliate resales. The key structural difference: Tier 1 carries no financial-statement assurance requirement. Both tiers are modeled on Regulation A with the same offering limits, and issuers would file on a new Form 1-CRYPTO.³ The architecture is a crypto-tailored Reg A analogue rather than a reinvention—which is exactly what the March remarks signalled, minus the tiering, which they did not mention.

What likely distinguishes it from Reg A+ Tier 2 is disclosure tailoring. Reg A+ requires Form 1-A with audited financials, ongoing reports (Forms 1-K, 1-SA, 1-U), and a four-to-six-month SEC qualification process. Crypto issuers running through Reg A+ today produce disclosure packages designed for Tier 2 retail securities, then bolt on token-economic, smart-contract-risk, and decentralization-roadmap disclosures because no native template exists. The fundraising exemption is the chance to fix this: structured schedules built around what crypto investors actually evaluate—token allocations, vesting, governance distribution, validator architecture, treasury management—rather than what a 1933-Act drafter would have included. The pathway needs Section 18 covered-security preemption to be a viable Reg A+ alternative; the speech’s explicit Reg A+ comparison suggests the SEC understands this.

What the proposal still leaves open. Whether the qualification process will be expedited (Reg A+‘s four-to-six-month window is the rail’s biggest weakness for crypto issuers; without an expedited track, the fundraising exemption is Reg A+ in a different costume); whether existing Reg A+ qualifications can convert without restarting; whether secondary trading will be permitted on registered exchanges or only on alternative trading systems; whether qualification staff will be experienced enough with token-specific disclosure to keep the timeline meaningful.

For founders raising between $5 million and $75 million, the fundraising exemption likely competes directly with Reg A+ Tier 2 and Reg D 506(c). Which wins depends on disclosure burden, qualification timeline, and whether the new pathway delivers cleaner secondary-trading mechanics.

The Investment Contract Safe Harbor: Status-Based Exit

The third pillar is the headline, and it is now proposed Rule 400. If its conditions are met, a crypto asset is deemed not to be subject to an investment contract for purposes of the definitions of “security” in both the Securities Act and the Exchange Act.³ The exit is status-based, not time-based; there is no four-year clock inside the safe harbor. The mechanism is a certification the issuer files with the Commission on a new form, stating in the first person that it has completed or permanently ceased all essential managerial efforts it promised or represented it would engage in, together with a narrative analysis supporting that certification.³

Atkins did not propose to declassify securities by administrative fiat. He proposed to operationalize Howey itself: SEC v. W.J. Howey Co., 328 U.S. 293 (1946), defines an investment contract as requiring an expectation of profit derived from the entrepreneurial or managerial efforts of others. The SEC’s theory is that when those essential managerial efforts cease, the fourth Howey prong dissolves and the investment-contract relationship factually terminates—an extrapolation from Howey the SEC will defend, but one no Supreme Court case has endorsed at the cessation step. The safe harbor provides bright-line evidentiary triggers—semi-annual disclosures on decentralization progress, governance distribution, validator counts, token allocation changes, material network events—for what the SEC argues is a doctrinal conclusion Howey already supports. The architecture tracks Peirce’s Proposed Rule 195 modified for status-based exit: an issuer that elects in accepts initial registration relief in exchange for ongoing disclosure cadence; antifraud liability under Section 17(a), Section 10(b), 15 U.S.C. § 78j(b), and Rule 10b-5 survives throughout.

The exit triggers when the issuer’s essential managerial efforts cease. The bright-line operational triggers most likely include completion of the represented network functionality, distribution of governance authority across a defined validator set, code immutability or community-governed upgrades with adequate time-locks, and absence of issuer control over treasury or fee distribution. The exit test is expected to align with the SEC/CFTC taxonomy’s three-factor digital-commodity test⁴—meaning a project satisfying the safe-harbor exit also satisfies the taxonomy’s commodity-classification criteria—though the alignment is not yet codified.

This is meaningfully different from how trade press has covered the framework. The colloquial “Token Safe Harbor 3.0” framing implies a categorical exemption that converts securities into non-securities by SEC rule. The actual proposal is doctrinally tighter: a Howey-faithful factual-cessation rule—at least on the SEC’s reading—that the SEC will defend under the framework Howey itself supplies. The distinction has material consequences for the rule’s litigation vulnerability, addressed in the Loper Bright challenge analysis below.

For founders building L1/L2 protocols or DeFi infrastructure with credible decentralization roadmaps, the safe harbor is the most consequential of the three exemptions. But the “status, not timer” architecture means founders cannot simply schedule the exit; they have to engineer the cessation.

Whether any of the three pathways would even apply to your project is a separate, gate-by-gate question—many tokens never needed an exemption at all after the March 2026 taxonomy. That eligibility analysis lives in the firm’s companion guide: who qualifies for the token exemption, which walks founders through the Three-Gate Eligibility Test. For sequencing the raise itself against the live rules, see the token launch legal checklist.

Which Tokens Did the SEC and CFTC Name as Digital Commodities?

Sixteen futures-based tokens, plus two additional examples. The March 2026 SEC-CFTC interpretive release names sixteen tokens as digital commodities—BTC, ETH, SOL, XRP, ADA, DOGE, and ten more—because each underlies CFTC-regulated futures, adds Algorand and LBRY Credits as non-futures examples, and presents the list as illustrative rather than exhaustive.⁴

A digital commodity, in the release’s architecture, is a crypto asset that is not itself a security—the holder’s return depends on market forces and network use, not on a promoter’s essential managerial efforts. The release gives the category its most concrete content in a single enumerated passage: “Examples of digital commodities include Aptos (APT); Avalanche (AVAX); Bitcoin (BTC); Bitcoin Cash (BCH); Cardano (ADA); Chainlink (LINK); Dogecoin (DOGE); Ether (ETH); Hedera (HBAR); Litecoin (LTC); Polkadot (DOT); Shiba Inu (SHIB); Solana (SOL); Stellar (XLM); Tezos (XTZ); and XRP (XRP).”⁴ The release separately names Algorand (ALGO) and LBRY Credits (LBC) as additional examples not grounded in futures listings—the futures rationale is illustrative, not a gate.

The table below is the firm’s canonical record of the named list. Every one of the sixteen futures-based tokens is named on the same basis—it underlies CFTC-regulated futures on a designated contract market—and carries the same practical consequence: spot transactions in the asset sit outside SEC securities jurisdiction. The table therefore records only the differentiating data; Algorand and LBRY Credits are the release’s two additional non-futures examples.

The release presents the list as illustrative, not exhaustive; absence from this table does not mean a token is a security, and presence does not immunize a fundraising transaction structured as an investment contract.

The release classifies the asset; it does not bless the transaction. A named digital commodity sold through a fundraising scheme that promises profits from the seller’s efforts is still an investment-contract offering under Howey, and an unnamed token whose network is genuinely functional can qualify as a digital commodity on the release’s own characteristics test. How the five categories work is the subject of the firm’s companion analysis of the SEC-CFTC five-category token taxonomy.

If your token appears in this table, your spot-market classification question is largely answered—but your fundraising-transaction question is not. Run the transaction analysis before you rely on the label.

What Replaced the SEC’s 2019 Token Framework?

The March 2026 interpretive release supersedes the SEC staff’s April 2019 Framework for “Investment Contract” Analysis of Digital Assets, replacing its multi-factor balancing exercise with a five-category taxonomy in which only digital securities face SEC registration.⁴

The 2019 Framework governed classification practice for nearly seven years—April 2019 to March 2026—and it was a list of dozens of weighing factors with no hierarchy, which in practice meant nearly any token could be argued into investment-contract status. The 2026 release replaces that with five defined categories—digital commodities, digital collectibles, digital tools, payment stablecoins, and digital securities—of which only digital securities face SEC registration. The release also ties classification to functionality: a crypto system is functional when its native asset can be used on the system in accordance with the system’s programmatic utility.

For founders the practical shift is directional: classification stopped being a facts-and-circumstances essay and became a category assignment with named examples. The full category-by-category analysis lives in how the five categories work; the point for this guide is that the taxonomy is the operative half of Regulation Crypto—in effect since March 23, 2026—while the exemption half is still only proposed.

How the Innovation Exemption Differs from Peirce’s Token Safe Harbor 1.0 and 2.0

Three differences separate Atkins’s framework from Commissioner Peirce’s Token Safe Harbor proposals: a status test instead of a three-year clock, Chair-driven Commission rulemaking instead of a single-Commissioner proposal, and coupling to the March 2026 taxonomy that gives the cessation test an objective classification anchor. Trade press has nonetheless treated the framework as “Token Safe Harbor 3.0”—a framing that overstates institutional continuity in ways that affect how confidently founders should rely on the rule. The lineage supplies the support.

Origins: Peirce’s Token Safe Harbor 1.0 and 2.0

The original proposal was Commissioner Hester Peirce’s Running on Empty speech of February 6, 2020, the package now colloquially called Token Safe Harbor 1.0.⁸ Peirce floated a three-year regulatory grace period during which network developers could distribute tokens, build infrastructure, and reach what she called Network Maturity—the point at which the network was either functional or sufficiently decentralized that Howey’s fourth prong no longer attached. Token Safe Harbor 2.0, released as Proposed Rule 195 on April 13, 2021, refined the architecture but kept the three-year window.⁹ Both proposals were single-Commissioner work product, never Commission action.

What Survived from Peirce’s Framework

What conceptually survived from Peirce’s framework into Atkins’s: the disclosure regime built around decentralization milestones, the exit test grounded in essential-managerial-efforts cessation, semi-annual disclosures during the runway, and antifraud preservation throughout. The doctrinal architecture—a runway-then-exit structure under which a network earns its way out of investment-contract status—is Peirce’s.

What Atkins Changed

What Atkins changed is more important than what he kept. Atkins’s safe harbor uses a status test, not Peirce’s three-year clock; the startup exemption is a separate pillar with its own four-year duration cap (which trade press has confused with the safe harbor’s nonexistent timer); the framework is coupled with the March 2026 SEC/CFTC five-category token taxonomy that gives the cessation test an objective classification anchor; and—for the first time in this lineage—a Chair is driving rulemaking through OIRA rather than circulating an aspirational policy statement. The institutional posture is fundamentally different.

The “3.0” framing is shorthand. The reality is that Atkins’s framework draws on Peirce’s work but is not Peirce’s proposal. Founders relying on the framework should source their understanding from Atkins’s March 17, 2026 speech, the eventual NPRM, and the SEC/CFTC interpretive release that frames the underlying classification regime⁴—not from Peirce 1.0/2.0, which never reached Commission action and which differ in important respects.

The side-by-side makes the differences concrete:

The Exemptive-Authority Challenge Risk: Loper Bright + Howey Fidelity

The framework’s most attractive feature—the Investment Contract Safe Harbor’s status-based exit—is also its most legally vulnerable. The elements of an APA challenge are already visible to sophisticated practitioners, and founders should understand the shape of that challenge before treating the safe harbor as a settled regulatory fixture.

Start with what the proposal did and did not do about its own authority. The Unified Agenda’s “Not Yet Determined” entry is superseded: the release now carries a statutory-authority recital invoking Securities Act sections 3(b), 18, 19(a), and 28 and Exchange Act sections 3(b), 12, 13, 23(a), and 36, and the authority citation for the new part 228 lists 15 U.S.C. 77c, 77r(b)(3), 77s, 77z-3, 78c(b), 78w, and 78mm.³ Section 28 is there, as this firm and others inferred it would be. What is still missing is an argument. The release explains its authority for the fundraising exemption in the discussion itself, invoking Section 28 by name; the safe-harbor discussion argues none, running instead on purpose—clarity, certainty, capital formation. A recital lists power. It does not explain why that power reaches this move. A challenger reads the difference, and the difference is where the case starts.

The cleanest version of the SEC’s defense is the doctrinal one described above: the safe harbor operationalizes Howey rather than declassifying securities by fiat, with the disclosure runway supplying evidentiary triggers for a conclusion Howey already supports. Section 28 of the Securities Act, 15 U.S.C. § 77z-3, provides express discretionary authority for the startup and fundraising exemptions during the runway, and even after Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), Section 28’s “necessary or appropriate in the public interest, and is consistent with the protection of investors” language preserves substantial agency discretion within its boundaries.

The cleanest version of the attack runs differently. Loper Bright eliminated Chevron deference for ambiguous statutes. Skidmore deference—which gives weight to agency interpretations only to the extent they are persuasive—survives, but it is meaningfully weaker than Chevron, and the SEC cannot count on it. Challengers will argue that “investment contract,” as defined in Section 2(a)(1) of the Securities Act and elaborated by Howey, sits outside Section 28’s delegated boundary—that Section 28 authorizes exemptive rulemaking but not redefinition of the underlying definitional category. The closest on-point precedent is Financial Planning Association v. SEC, 482 F.3d 481 (D.C. Cir. 2007), in which the D.C. Circuit vacated an SEC rule that used general rulemaking authority to redefine a statutory category Congress had already addressed. The FPA court’s framing—quoting the Supreme Court’s holding in Board of Governors of the Federal Reserve System v. Dimension Financial Corp.—was direct: “the [SEC] has no power to correct flaws that it perceives in the statute it is empowered to administer.”¹⁰ A second front of attack is that the safe harbor’s bright-line proxies are themselves contestable as SEC-imposed rulemaking that goes beyond Howey’s factual test. Add the major-questions doctrine overlay—the proposition that resolving the regulatory status of a multi-trillion-dollar asset class is the paradigmatic “extraordinary case[]” demanding explicit Congressional authorization, West Virginia v. EPA, 142 S. Ct. 2587 (2022)¹¹—and the safe harbor’s vulnerability to APA challenge is real.

A defender will distinguish FPA on its facts, and the distinction has real force. FPA invalidated a rule that redefined a statutory exclusion Congress had drawn—the broker-dealer exclusion from “investment adviser”—whereas the Innovation Exemption leaves “investment contract” as Section 2(a)(1) defines it and provides exemptive relief under Section 28’s express delegation. Exemption is doctrinally distinct from redefinition. The rebuttal is that the safe harbor’s bright-line proxies—validator counts, governance analytics, treasury protocols—effectively redefine when Howey’s fourth prong attaches and dissolves. Whether that operates as exemption (Section 28 territory) or as definitional gloss (outside Section 28) is the litigated question. Both sides have real arguments. This article’s thesis does not turn on which wins; it turns on the litigation being live.

The SEC will not be without responses: Section 28 has been on the books since 1996, used dozens of times for exemptive rulemakings including Reg A+‘s 2015 expansion, and a defender will argue the Innovation Exemption is routine exemptive rulemaking—precisely the clarification Congress invited—not an unprecedented assertion of authority. Reasonable judges will disagree.

None of this means founders should ignore the safe harbor. When the rule is finalized, compliance will be necessary. But it does mean compliance is necessary-but-not-sufficient. Until the rule clears APA challenge in a circuit court—and recent Fifth Circuit jurisprudence vacating SEC action on statutory-authority grounds is unmistakable: Nat’l Ass’n of Private Fund Managers v. SEC, 103 F.4th 1097 (5th Cir. 2024) (vacating the SEC’s private fund advisers rule in full); see also All. for Fair Bd. Recruitment v. SEC, 125 F.4th 159 (5th Cir. 2024) (en banc) (9-8 decision vacating SEC’s approval of Nasdaq board diversity rules; invoking major-questions doctrine and holding the SEC failed to identify clear congressional authorization for rules of “staggering” economic and political significance), together with the D.C. Circuit’s FPA v. SEC precedent—founders should structure offerings as if the safe harbor might be vacated. The Alliance en banc opinion is particularly relevant: a 9-8 majority found the SEC’s statutory-authority showing inadequate for a rule far less aggressive than redefining when a Howey fourth prong dissolves. The mitigation discipline—stated in full in the cessation-failure analysis below—is preserved live-exemption compliance, Howey-rigor documentation, and registration optionality. Statutory clarification through the CLARITY Act framework, if enacted, would moot most of these concerns.

The most coherent crypto-securities framework Washington has produced may also be its most vulnerable. Treat the Investment Contract Safe Harbor as a probabilistic compliance vehicle, not regulatory bedrock—and structure offerings to survive without it.

The Decision Matrix: Innovation Exemption vs. Reg D vs. Reg S vs. Reg A+ vs. Wait-for-CLARITY

A founder making capital decisions in mid-2026 has five live pathways. Each carries a different process burden, timeline, investor pool, marketing rules, secondary-market posture, and—critically—state-law preemption status. The burden differential between covered-security pathways (single-track federal compliance) and non-preempted pathways (fifty-state blue-sky filings) is often the deciding factor for mid-size raises, not the cap or timeline.

*Innovation Exemption pathways are proposed, not available. Conditions are drawn from the August 18, 2026 proposing release and may change before adoption; nothing may be offered or sold in reliance on them until a final rule takes effect. Reg D 506(b)/(c) holding periods are 12 months for non-reporting issuers, 6 months for Exchange Act reporting issuers under Rule 144.

Reg D 506(c) wins on speed when the investor pool is sophisticated. The four-to-eight-week timeline, federal preemption under Section 18(b)(4)(F) of the Securities Act, 15 U.S.C. § 77r(b)(4)(F), and public-marketing permission make 506(c) the dominant rail for tokens raising from accredited investors. The trade-off is the twelve-month holding period and secondary-trading restrictions; both compress the liquidity profile founders can offer.

Reg A+ Tier 2 wins on retail liquidity. The $75 million cap, the audited-financials and ongoing-reporting burden, and the four-to-six-month qualification timeline are the price of (a) selling to non-accredited investors, (b) immediate secondary-trading permission, and (c) covered-security preemption under Section 18(b)(3) of the Securities Act, 15 U.S.C. § 77r(b)(3)—the qualified-purchaser pathway, not the (b)(4) statutorily-enumerated pathway some practitioners conflate.

The Innovation Exemption pathways win on the dimensions Reg D and Reg A+ leave open, but only if the NPRM delivers what Atkins’s speech promised. The Startup tier likely wins for early-stage projects under $5 million if the rule delivers covered-security status and principles-based disclosure—materially cheaper than Reg D 506(c)‘s verification regime. The Fundraising tier likely competes with Reg A+ Tier 2 for mid-size raises if the qualification track is genuinely expedited and the disclosure schedule is crypto-tailored; if not, it adds little over Reg A+. The Safe Harbor wins for tokens with credible decentralization roadmaps that can engineer essential-managerial-efforts cessation—carrying the legal-vulnerability discount from the Loper Bright challenge analysis above.

Waiting for the CLARITY Act is the conservative play. The Tillis-Alsobrooks compromise of May 1-2, 2026 restored momentum after the stablecoin-yield deadlock,¹² but the White House’s July 4, 2026 passage target came and went without enactment, and the Act remains pending as of this writing. If CLARITY passes, the Innovation Exemption framework becomes statutory, the legal-vulnerability discount disappears, and the regulatory architecture settles. Banking-industry pushback makes the timeline uncertain.¹³ Projects that can defer six-to-twelve months without competitive consequence may wait; most cannot.

Founders raising under $5 million should compare Innovation Exemption Startup against Reg D 506(b)/(c) on the proposed terms now, and again against the final rule. Founders raising $5-75 million should run the same comparison against Reg A+ Tier 2, tier by tier. Founders building decentralization-track protocols should compare the Safe Harbor against waiting for CLARITY—but should not assume the Safe Harbor is litigation-proof.

Token-Archetype Mapping: Which Pathway Fits Your Project

The decision matrix is the abstract version. The decision tree below maps the matrix to the token archetypes founders actually build. Use the same archetype framework as the SEC/CFTC token taxonomy’s five categories⁴—the classification analysis sits upstream of every exemption decision.

Governance Token Under $5 Million Raise

If the token confers governance rights without revenue distribution and the launch capital fits within $5 million, the Startup Exemption is the likely best fit—assuming the NPRM delivers covered-security preemption. Reg D 506(c)‘s accredited-only verification regime is overkill for governance tokens that need broad community distribution to function as governance tokens. But governance tokens that distribute revenue or confer claims on a treasury asset pool may be classified as digital securities under the taxonomy, in which case the Startup Exemption’s principles-based disclosure may not provide adequate investor protection, and the project should default to Reg D 506(c) or the Fundraising tier even at sub-$5-million scale. The classification analysis is upstream; do not skip it. Cross-reference: SEC/CFTC token taxonomy, five-category framework.

Utility Token, $20 Million to $50 Million Raise

If the token provides genuine network utility with a clear functional purpose at launch and the raise sits between $20 million and $50 million, the comparison is between Reg A+ Tier 2 and the Fundraising tier. Reg A+ wins today on certainty—the rail is live, the disclosure framework is settled, the qualification timeline is predictable. The Fundraising tier wins if the NPRM delivers an expedited qualification track and a disclosure schedule built around token-specific risks. The conservative call: start Reg A+ Tier 2 preparation in parallel with monitoring NPRM developments, convert to the Fundraising tier if the rule publishes with favorable mechanics, default to Reg A+ if not. Dual-tracking the early disclosure work is modest cost relative to the option value.

L1/L2 Token with Multi-Year Decentralization Plan

The Investment Contract Safe Harbor is the natural fit—and the most legally fraught. Layer-1 and Layer-2 protocol tokens have credible paths to essential-managerial-efforts cessation: validator distribution, governance handoff to community-controlled DAOs, code immutability with time-locked upgrades, and absence of issuer control over treasury or fee distribution. Documenting the decentralization milestones to Howey-rigor levels, while running semi-annual disclosure cadence, is the operational discipline the safe harbor will reward.

But the Loper Bright + Howey-fidelity vulnerability analyzed above applies most acutely here. The L1/L2 founder is making a multi-year capital and architecture commitment on a rule that has not survived APA challenge, and the mitigation discipline in the cessation-failure analysis below applies in full. Treat the safe harbor as the most likely path; treat alternatives as fallback architecture.

The Cessation-Failure Path: What If Your Network Never Achieves Cessation

The Investment Contract Safe Harbor’s most attractive feature is also its most dangerous. The failure scenarios are not edge cases. They are the foreseeable outcome for the meaningful subset of L1/L2 projects whose teams continue making material decisions about protocol direction, fee distribution, and treasury management long past the runway.

If the cessation test is never satisfied—because the founding team continues to drive material network decisions, or because the network’s operational reality remains centralized despite token-distribution choreography—the safe harbor never triggers. The token remains an investment contract. The project does not default to a clean status; it defaults to enforcement uncertainty for an unregistered securities offering.

Section 12(a)(1) of the Securities Act, 15 U.S.C. § 77l(a)(1), provides purchasers of unregistered, non-exempt securities a strict-liability rescission claim. Section 13, 15 U.S.C. § 77m, sets a one-year limitations period for Section 12(a)(1) claims (running from the violation, with no discovery rule) and a three-year statute of repose running from the date the security was bona fide offered to the public. Whether the safe-harbor pendency tolls the limitations clock will be a litigated question; founders should not assume tolling. The repose is not subject to equitable extension. Cal. Pub. Emps.’ Ret. Sys. v. ANZ Sec., Inc., 137 S. Ct. 2042 (2017).

The Wells-notice posture compounds the exposure. SEC enforcement counsel will tee up Wells notices when issuers withdraw from safe-harbor compliance without satisfying the cessation test, and the project’s prior reliance on the safe harbor becomes record evidence of awareness of the registration requirement—which can support scienter findings on related Section 17(a) and Rule 10b-5 claims. The mechanic is not hypothetical. Cf. SEC v. LBRY, Inc., No. 1:21-cv-00260, 2022 WL 16744741 (D.N.H. Nov. 7, 2022) (rejecting issuer’s reliance on the absence of an ICO as a safe harbor against Section 5 liability and holding that the issuer’s own statements and internal documents supplied adequate evidence that token purchasers were led to expect profits from the issuer’s efforts). LBRY’s regulatory-positioning theory was different from a safe-harbor election, but the doctrinal mechanic is the same: pre-enforcement regulatory posture becomes record evidence at the merits stage. The defense framework the firm has previously analyzed in SEC Crypto Enforcement Defense: What to Do When You Get a Wells Notice applies to this scenario, with the additional complication that the project’s safe-harbor election sits in the SEC’s file.

The Loper Bright vacatur scenario is the catastrophic version. If the safe harbor itself is invalidated on APA review, founders who structured offerings around the rule face retroactive Howey exposure on the original offering and every subsequent transfer, with Section 12(a)(1) rescission claims live for the full repose window.

The taxonomy mismatch risk adds a third failure mode. The safe-harbor exit test is expected to align with the SEC/CFTC taxonomy’s three-factor digital-commodity test⁴—but the alignment is not codified as of OIRA review. A project could satisfy the cessation test without satisfying the taxonomy’s digital-commodity criteria, leaving post-cessation status ambiguous. Or vice versa. Founders should not assume alignment until the NPRM confirms it.

The mitigation discipline is fourfold. Preserve Reg D 506(c) compliance during the runway so a cessation failure does not collapse into a Howey-only defense. Document decentralization to Howey-rigor levels—validator metrics, governance analytics, code-immutability commitments, treasury protocols—not only to the safe harbor’s bright-line proxies. Build Reg A+ qualification optionality into the cap table. Track Section 13 repose clocks per offering tranche; the three-year repose is the operational planning horizon for catastrophic exposure.

Plan for cessation success and cessation failure from year zero—and build the offering to survive without the safe harbor.

DeFi Protocol Token

DeFi protocol tokens are the hardest case. The CLARITY Act’s three-factor DeFi test (no unilateral control over user assets, immutable or community-governed software, meaningful decentralization) maps cleanly onto the safe harbor’s cessation criteria. But the doctrinal vulnerability runs deepest here—DeFi protocols are precisely the asset class for which the SEC’s “essential managerial efforts cease” theory is most contestable, because critics argue that protocol-level managerial decisions persist in upgrade authority, parameter setting, and frontend control even after nominal “decentralization.” Cross-reference: The CLARITY Act Explained: CFTC vs. SEC Jurisdiction Defined.

The conservative posture for DeFi protocols is to combine the Safe Harbor with the most rigorous decentralization architecture available—non-custodial design, immutable smart contracts or multi-year time-locked governance, fully open-source code with distributed development, absence of upgrade keys. Safe Harbor compliance plus the strongest possible Howey-cessation factual record gives the project the best chance of surviving APA challenge if the rule is invalidated.

Out-of-Scope Archetypes: Memecoins and Stablecoins

Two archetypes generally sit outside the Innovation Exemption’s exemption-selection process. Memecoins and community tokens often fall under the taxonomy’s digital-collectible or digital-tool categories⁴—not securities under Howey, no exemption needed—though the trap is that price-appreciation marketing, team-driven roadmaps, or revenue-sharing tokenomics can collapse the analysis back into investment-contract status, in which case the Startup Exemption’s principles-based disclosure is the most likely fit. Stablecoins follow the GENIUS Act Permitted Payment Stablecoin Issuer framework, Pub. L. No. 119-27, 139 Stat. 419 (2025): PPSI-issued payment stablecoins are statutorily defined as not securities under federal law and not commodities under the Commodity Exchange Act, 7 U.S.C. § 1 et seq. The exception is yield-bearing and algorithmic stablecoins, which fall outside PPSI and may be classified as digital securities—the Tillis-Alsobrooks compromise of May 1-2, 2026 attempted to draw the line between permitted “bona fide activities” rewards and prohibited bank-deposit-equivalent yield, but whether that line survives the final CLARITY Act text remains uncertain. Cross-reference: GENIUS Act Stablecoin Compliance Roadmap; Federal vs. State Stablecoin Regulation.

What’s Still Unknown (Honest Open Questions)

The proposal answered several of the questions this guide has carried since May—most importantly blue-sky preemption, the tiering of the fundraising exemption, and the statutory authority the agenda left blank. What follows is what it did not answer, plus what a proposal cannot answer because only adoption and litigation can. Verify every requirement above against the release text itself, and re-verify it against the final rule if one is adopted:

What the proposal does not do is argue its own authority for the safe harbor. It recites authority—Securities Act sections 3(b), 18, 19(a), and 28, Exchange Act sections 3(b), 12, 13, 23(a), and 36—and it argues Section 28 in the discussion for the fundraising exemption. For Rule 400 it argues nothing, and it never addresses whether a definitional safe harbor is exemptive relief under Section 28 or definitional rulemaking under Section 19(a) and Exchange Act Section 3(b). That framing question was always going to be the swing variable in the Loper Bright analysis above; the proposal declines to pick a side, which leaves the challenger to pick one for it.³

AML and KYC integration remains TBD. The Bank Secrecy Act framework imposes obligations on money services businesses, and the GENIUS Act, Pub. L. No. 119-27, 139 Stat. 419 (2025), brings Permitted Payment Stablecoin Issuers into the BSA framework. Whether the Innovation Exemption pathways will impose comparable obligations, or leave AML/KYC to FinCEN coordination, will determine whether the exemptions are operationally cheaper or more expensive than current Reg D and Reg A+ practice.

State blue-sky preemption is answered, and favorably. This guide called it the most consequential operational question for mid-size raises, and the proposal resolves it through NSMIA’s covered-security framework: a new definition of “qualified purchaser” under Section 18(b)(3) of the Securities Act, 15 U.S.C. § 77r, preempting state registration and qualification for offerings under the regulation and for certain secondary transactions in the same asset.³ Two cautions survive. Preemption follows the exemption, so an offering that falls outside the exemption loses preemption with it. And the secondary-market preemption is conditioned—current reporting is its price, which is precisely what filing the safe-harbor certification would end. Read those two together before treating either as free.

SEC-CFTC coordination on safe-harbor exit alignment with digital-commodity classification is TBD. The joint interpretive release adopted March 17, 2026⁴ established the five-category framework, but whether the cessation test is treated as automatic graduation into digital-commodity status, or whether the CFTC retains authority to challenge the classification, will affect post-cessation regulatory architecture.

The Loper Bright + Howey-fidelity attack discussed in the challenge-risk analysis above is the most consequential litigation-exposed open question. Until the rule is tested in a circuit court—recent Fifth Circuit jurisprudence vacating SEC action on statutory-authority and major-questions grounds, Nat’l Ass’n of Private Fund Managers, 103 F.4th 1097, and All. for Fair Bd. Recruitment v. SEC, 125 F.4th 159 (5th Cir. 2024) (en banc), together with the D.C. Circuit’s Financial Planning Association v. SEC, 482 F.3d 481 (D.C. Cir. 2007), precedent, all signal active scrutiny—founders should structure offerings as if the safe harbor might be invalidated.

Banking-industry pushback adds a parallel uncertainty vector. The CLARITY Act stablecoin-yield compromise drew crypto-industry support but is provoking opposition from traditional banks concerned about deposit substitution.¹² Similar pressure could reach the Innovation Exemption rulemaking—particularly the Fundraising Exemption’s positioning against bank-eligible offerings. Political durability across the next administration transition compounds the legislative uncertainty.

Cross-border issuance treatment remains undefined. Whether a Singapore Pte. Ltd. or Cayman SPV can use the Innovation Exemption pathways for U.S. distributions, and how the rule interacts with Regulation S’s offshore safe harbor, are not addressed in Atkins’s speech and may not be addressed in the NPRM. International issuers with U.S. token-purchaser populations should not assume accommodation.

Retroactive coverage for tokens already issued is uncertain. Whether founders who launched under SAFT structures, Reg D 506(b) private placements, or unregistered offerings can use the Innovation Exemption to remediate prior compliance gaps is not addressed in the speech. The default assumption—prospective application only—is conservative; whether the NPRM provides a transition or retroactive election will materially affect projects with pre-2026 offerings on their cap tables.

The framework is directional, not final. Track NPRM publication; submit specific, solution-oriented comment letters; preserve flexibility in offering documents drafted before the rule publishes.

When Could Regulation Crypto Actually Take Effect?

Mid-2027 at the earliest. A 2026 proposal opens a sixty-to-ninety-day comment period under the Administrative Procedure Act, followed by adoption, an effective date, and probable litigation—founders planning 2026 raises cannot wait for this rule.¹⁴

The sequence is fixed by 5 U.S.C. § 553: notice, comment, consideration of comments, final rule, effective date. Each node below is labeled for what it is—fact, target, or projection:

• March 17, 2026—Atkins announces the framework (fact).¹

• March 23, 2026—SEC-CFTC interpretive release published and effective, 91 Fed. Reg. 13,714 (fact).⁴

• April 6, 2026—rulemaking package reported submitted to OIRA (press-reported date; still under OIRA review per July 7, 2026 reporting).²

• April 2026—agenda’s NPRM target month; passed without publication (fact).¹⁵

• July 2026—press-reported revised NPRM target (target, press-sourced).¹⁶

• Early August 2026—extended OIRA review window exhausts under EO 12866 practice (computed from the reported April 6 submission).

• August 13, 2026—Sunshine Act notice published setting an open meeting to consider proposing the offering regime, 91 Fed. Reg. 52,392 (fact).¹⁷

• August 14, 2026—noticed meeting date; no proposing release issued that day (fact).¹⁸

• August 18, 2026—Regulation Crypto Assets proposed, Release Nos. 33-11434 and 34-106150, File No. S7-2026-27, 402 pages (fact).³

• Federal Register publication—not yet occurred as of August 19, 2026; the 60-day comment period runs from this date, not from the SEC’s announcement (pending).⁶

• 60 days post-publication—comment deadline, per the release’s own DATES section.³

• Mid-2027—earliest realistic effective date for a final rule (projection, per trade-press estimates and standard rulemaking timelines; litigation could extend it years).¹⁴

Every date after August 19, 2026 in that strip is a target or a projection, not a commitment. Build the raise on the live stack and treat adoption as upside.

What Founders Should Do Now

The Innovation Exemption is proposed, not in force. That distinction is the whole of the advice below: you can now read the conditions you will eventually have to meet, which makes structuring for them cheap, and you still cannot rely on any of them. Founders who structure offerings now against the actual rule text will be ready to elect in on day one. Founders who wait for a final rule will be a year late.

1. Run the Classification Analysis Under the Taxonomy First

If the token is a digital tool, digital collectible, or PPSI-issued payment stablecoin under the SEC/CFTC five-category taxonomy⁴, no exemption is needed—the token is not a security. The Innovation Exemption pathways are irrelevant. Use the Token Classification Issue Spotter to run the upstream classification analysis before evaluating any exemption. Skipping this step routes founders toward exemptions they may not need.

2. Map the Raise to the Comparison Matrix

For tokens classified as digital securities or sold through investment-contract structures, identify the top two pathways from the decision matrix above by cap, investor pool, marketing posture, and timeline. The top two will almost always be (a) the Innovation Exemption pathway aligned with the project’s scale, and (b) the live exemption (Reg D 506(c) or Reg A+ Tier 2) that serves as fallback if the rule is adopted unfavorably, is not adopted, or is invalidated on APA review. Plan for both.

3. Build Offering Documents With Optionality

Structure today’s offering documents so the same raise can convert into Innovation Exemption registration when the NPRM ships, or default to the existing exemption pathway if the rule publishes unfavorably. The dual-track work is not free, but the option value is high. Building Reg D 506(c) compliance materials in parallel with monitoring the NPRM is modest cost relative to restructuring an in-flight raise. For projects at $5-million-and-under, dual-tracking Reg D 506(b) and Innovation Exemption Startup is the most defensible posture.

4. Documentation Discipline From Year Zero

Document decentralization milestones, governance distribution, validator counts, code-immutability commitments, and treasury management protocols from first token sale. The cessation test will reward Howey-rigor factual records; the Loper Bright vacatur scenario will require them. Founders who treat decentralization as a milestone-driven engineering effort with documented evidence will have defensible records when the cessation test or the APA challenge arrives. Founders who treat decentralization as marketing language will not.

5. Comment—The Window Is Open Now

This is no longer a step to prepare for. The proposal is out, it asks 154 separate questions, and comments are due sixty days after Federal Register publication. Submit specific, solution-oriented letters citing the numbered questions—detailed examples of how a proposed mechanic fails for an actual token architecture, alternatives that keep investor protection while improving operational viability, demonstrations of industry practice in use. Generic “the framework is welcome” comments are useless. Practitioner-grade submissions get incorporated into final guidance. The crypto industry left the 2018-2024 enforcement era with weaker authorities than it should have because comment-period engagement was thin. The Innovation Exemption is the chance to fix that.

6. Disclosure Discipline—Antifraud Liability Survives Every Safe Harbor

The Innovation Exemption exempts registration. It does not exempt antifraud liability. Section 17(a) of the Securities Act, 15 U.S.C. § 77q, applies to every offer or sale of any security—including offers and sales inside the safe harbor’s runway. Section 10(b) of the Exchange Act, 15 U.S.C. § 78j(b), and Rule 10b-5 apply to every secondary-market transaction. Misleading whitepapers, exaggerated tokenomics projections, undisclosed conflicts of interest, and overstated decentralization claims remain actionable—by the SEC and by private plaintiffs. The safe harbor does not protect against any of this.

Founders who treat the Innovation Exemption as a license to relax disclosure will be the test cases for the rule’s antifraud preservation. The discipline is straightforward: every disclosure materially accurate when made; every projection reasonably grounded; every conflict disclosed; every milestone claim documented. The Innovation Exemption raises the operational floor for compliant token offerings; it does not raise the ceiling on good-faith disclosure.

The Wells-notice generation of crypto founders learned the cost of waiting for guidance. The Innovation Exemption generation should not repeat the mistake. Build for the rule that is coming, not the rule that exists—but build with disclosure discipline that survives every safe harbor and every regulatory regime.

Conclusion: The Most Coherent Federal Framework Yet—and the Risks

The taxonomy classified the assets. The Innovation Exemption, if it ships in recognizable form, gives them a registration pathway. The CLARITY Act, if it passes, gives them a jurisdictional home. Together, these three pieces represent the most coherent federal crypto framework the United States has ever proposed—the institutional architecture the industry has been requesting since the SEC’s 2017 declaration of jurisdiction.

Each piece is provisional. The taxonomy is an interpretive release subject to revision under a future administration. The Innovation Exemption is a proposal—402 pages of it, and no more binding for the page count—with credible APA-challenge exposure under Loper Bright and Financial Planning Association v. SEC, and a statutory authority the release recites without ever arguing for its central move. The CLARITY Act is pending legislation with active banking-industry pushback. The framework is real; durability is contingent.

The founders who win this transition will not wait for final rules. They will structure today’s offerings to elect into the framework on day one if it ships favorably, default to the existing exemption stack if it does not, and survive APA challenge or political reversal if either occurs. Documentation discipline, optionality, comment-period engagement, disclosure rigor—these translate regulatory transition into competitive advantage.

The founders who lose will treat a proposing release as the final rule, structure offerings as if the safe harbor were litigation-proof, and build decentralization to bright-line proxies rather than to Howey-rigor evidence. The framework is the most coherent the U.S. has produced. It is not bedrock yet—and structuring as if it were is how Wells notices get teed up two years from now.

How Astraea Counsel Helps

Astraea Counsel advises crypto founders on token classification and exemption-pathway selection. Engagement scope typically includes regulatory architecture design across the Reg D, Reg S, Reg A+, and Innovation Exemption stack; offering-document optionality drafting (dual-track Reg D 506(c) plus Innovation Exemption Startup, or parallel Reg A+ qualification); decentralization-milestone documentation to Howey-rigor levels; comment letter drafting for the NPRM record; and Wells Notice defense and pre-Wells posture for offerings sitting at enforcement risk. The firm publishes ongoing analysis of SEC digital-asset rulemaking—see the SEC Crypto Pivot guide, Token Launch Legal Checklist, and Wells Notice defense framework. Explore our Digital Assets & Blockchain services.

Related Resources

• Who Qualifies for the SEC’s Token Startup Exemption and Safe Harbor?—the gate-by-gate eligibility companion to this guide

• The SEC/CFTC Token Taxonomy: What the Five Categories Mean for Your Token—the classification framework this exemption sits on top of

• Crypto Enforcement Tracker (2024-2026)—the firm’s action-by-action dataset behind the enforcement-pace analysis above

• The CLARITY Act Explained: CFTC vs. SEC Jurisdiction Finally Defined—the legislative companion piece

• Token Launch Legal Checklist: Avoiding SEC Enforcement in 2025—the existing exemption pathways the Innovation Exemption competes with

• The SEC’s Crypto Pivot: What the Dismissals and Task Force Mean for Your Startup—the political shift that produced this framework

• Token Classification Issue Spotter—interactive tool to identify your token’s category

---

Disclaimer: This article provides general information for educational purposes only and does not constitute legal advice. As of August 19, 2026 the Innovation Exemption framework exists as a proposing release only, issued August 18, 2026 and not yet published in the Federal Register; descriptions of its requirements are sourced to that release, and a proposal confers no exemption and may be changed or abandoned before adoption. Final rule text may differ. Consult qualified legal counsel for advice on your specific offering. Attorney Advertising.

Footnotes

1. Paul S. Atkins, Chairman, U.S. Sec. & Exch. Comm’n, Regulation Crypto Assets: A Token Safe Harbor, Remarks at the DC Blockchain Summit (Mar. 17, 2026), https://www.sec.gov/newsroom/speeches-statements/atkins-remarks-regulation-crypto-assets-031726. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

2. The April 6, 2026 OIRA-submission date is press-reported; no OIRA docket record confirming the submission date was publicly available as of July 18, 2026. Review status corroborated by Jesse Hamilton, U.S. SEC to Propose Crypto Rule as Soon as This Month, CoinDesk (July 7, 2026) (the proposal “still under review at the White House Office of Information and Regulatory Affairs”), https://www.coindesk.com/policy/2026/07/07/u-s-sec-to-propose-crypto-rule-as-soon-as-this-month-to-ease-startups-fundraising; reginfo.gov, RIN 3235-AN38 (2026 Unified Agenda edition, Proposed Rule Stage). ↩ ↩²

3. Regulation Crypto Assets, Securities Act Release No. 33-11434, Exchange Act Release No. 34-106150, File No. S7-2026-27, RIN 3235-AN38 (proposed Aug. 18, 2026) (402 pages; proposing a new 17 C.F.R. part 228), https://www.sec.gov/files/rules/proposed/2026/33-11434.pdf. Subpart structure at 29-30; startup exemption at proposed 17 C.F.R. § 228.200; fundraising exemption tiers and affiliate sublimits at proposed § 228.300(a) and release 107; investment contract safe harbor at proposed § 228.400; “qualified purchaser” definition and preemption at release 30 and 170 (secondary-market preemption continuing only while the issuer satisfies the exemption’s disclosure, filing and periodic-reporting requirements); statutory authority recital at release 319; part 228 authority citation (15 U.S.C. 77c, 77r(b)(3), 77s, 77z-3, 78c(b), 78w, and 78mm) at release 322; comment period in the release’s DATES section. See also Press Release, U.S. Sec. & Exch. Comm’n, SEC Proposes New Regulation Crypto Assets, No. 2026-76 (Aug. 18, 2026), https://www.sec.gov/newsroom/press-releases/2026-76. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰ ↩²¹ ↩²²

4. Application of the Federal Securities Laws to Certain Types of Crypto Assets and Certain Transactions Involving Crypto Assets, Securities Act Release No. 33-11412, Exchange Act Release No. 34-105020, 91 Fed. Reg. 13,714 (Mar. 23, 2026), https://www.federalregister.gov/documents/2026/03/23/2026-05635/application-of-the-federal-securities-laws-to-certain-types-of-crypto-assets-and-certain. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹²

5. See, e.g., Sullivan & Cromwell LLP, SEC and CFTC Issue Interpretation Clarifying Application of Securities Laws to Crypto Assets (Mar. 2026), https://www.sullcrom.com/insights/memo/2026/March/SEC-Clarifies-Application-Securities-Laws-Crypto-Assets; Sidley Austin LLP, SEC Releases Landmark Interpretation on Application of U.S. Securities Laws to Crypto Assets (Mar. 24, 2026), https://www.sidley.com/en/insights/newsupdates/2026/03/sec-releases-landmark-interpretation-on-application-of-us-securities-laws-to-crypto-assets; Davis Polk & Wardwell LLP, SEC Begins to Clarify Application of Federal Securities Laws to Crypto (Mar. 24, 2026), https://www.davispolk.com/insights/client-update/sec-begins-clarify-application-federal-securities-laws-crypto. ↩

6. Astraea Counsel verification, August 19, 2026, against the Federal Register API record of every SEC document published August 14 through August 19, 2026 (45 documents, none of them a proposed rule and none bearing File No. S7-2026-27), https://www.federalregister.gov/api/v1/documents.json?conditions[agencies][]=securities-and-exchange-commission&conditions[publication_date][gte]=2026-08-14. The release’s own DATES section carries bracketed placeholders for both the publication date and the comment deadline, which is the form a proposing release takes before Federal Register publication. ↩ ↩²

7. The Industry Spread, SEC Puts Regulation Crypto on July Agenda with Three Rules (July 9, 2026), https://theindustryspread.com/sec-regulation-crypto-july-2026-agenda-three-rules/. ↩ ↩²

8. Hester M. Peirce, Comm’r, U.S. Sec. & Exch. Comm’n, Running on Empty: A Proposal to Fill the Gap Between Regulation and Decentralization, Remarks Before the Int’l Blockchain Cong. (Feb. 6, 2020), https://www.sec.gov/newsroom/speeches-statements/peirce-remarks-blockress-2020-02-06. ↩

9. Hester M. Peirce, Comm’r, U.S. Sec. & Exch. Comm’n, Token Safe Harbor Proposal 2.0 (Apr. 13, 2021), https://www.sec.gov/newsroom/speeches-statements/peirce-statement-token-safe-harbor-proposal-20. ↩

10. Fin. Planning Ass’n v. SEC, 482 F.3d 481, 491 (D.C. Cir. 2007) (“the [SEC] has no power to correct flaws that it perceives in the statute it is empowered to administer”) (quoting Bd. of Governors of Fed. Reserve Sys. v. Dimension Fin. Corp., 474 U.S. 361, 374 (1986)). ↩

11. West Virginia v. EPA, 142 S. Ct. 2587, 2609 (2022) (major-questions doctrine; “extraordinary cases” framing). ↩

12. See Nikhilesh De & Jesse Hamilton, CLARITY Act Text Lets Crypto Firms Offer Stablecoin Rewards While Shielding Bank Yield, CoinDesk (May 1, 2026), https://www.coindesk.com/policy/2026/05/01/clarity-act-text-lets-crypto-firms-offer-stablecoin-rewards-while-shielding-bank-yield; Francisco Rodrigues, Crypto Industry Backs CLARITY Act Yield Compromise, Pushes Senate Banking for Markup, CoinDesk (May 2, 2026), https://www.coindesk.com/policy/2026/05/02/crypto-industry-backs-clarity-act-yield-compromise-pushes-senate-banking-for-markup; see also Zack Abrams, Coinbase Says Deal Reached on CLARITY Act Stablecoin Yield, Clearing Path to Long-Stalled Senate Markup, The Block (May 2, 2026), https://www.theblock.co/post/399780/coinbase-says-deal-reached-on-clarity-act-stablecoin-yield-clearing-path-to-long-stalled-senate-markup. ↩ ↩²

13. See Crypto Times, Lummis and Tillis Defend CLARITY Act Stablecoin Compromise as Banking Lobby Mounts Pushback (May 5, 2026), https://www.cryptotimes.io/2026/05/05/lummis-and-tillis-defend-clarity-act-stablecoin-compromise-as-banking-lobby-mounts-pushback/ (joint statement from American Bankers Association, Bank Policy Institute, Consumer Bankers Association, Financial Services Forum, and Independent Community Bankers of America criticizing the Tillis-Alsobrooks compromise). ↩

14. 5 U.S.C. § 553 (APA rulemaking procedure); see also The Cryptonomist, SEC Crypto Regulation 2026 (July 11, 2026), https://en.cryptonomist.ch/2026/07/11/sec-crypto-regulation-2026/ (projecting a mid-2027 earliest effective date; cited as an estimate, not a commitment). ↩ ↩²

15. Office of Information and Regulatory Affairs, Unified Agenda of Regulatory and Deregulatory Actions, Crypto Assets, RIN 3235-AN38 (SEC, Division of Corporation Finance) (Proposed Rule Stage; Legal Authority “Not Yet Determined”; the Spring 2025 agenda edition (pubId 202504) carried an NPRM timetable of 04/00/2026, and the current 2026 edition (pubId 202510, published July 7, 2026) carries 07/00/2026), https://www.reginfo.gov/public/do/eAgendaViewRule?pubId=202510&RIN=3235-AN38. ↩

16. CoinDesk, U.S. SEC to Propose Crypto Rule as Soon as This Month to Ease Startups’ Fundraising (July 7, 2026), https://www.coindesk.com/policy/2026/07/07/u-s-sec-to-propose-crypto-rule-as-soon-as-this-month-to-ease-startups-fundraising. ↩

17. Securities and Exchange Commission, Sunshine Act Meetings, 91 Fed. Reg. 52,392 (Aug. 13, 2026) (FR Doc. 2026-16519; dated Aug. 10, 2026; filed Aug. 11, 2026 at 4:15 p.m.) (noticing an Open Meeting on Friday, August 14, 2026 at 10:00 a.m. ET at which “[t]he Commission will consider whether to issue a release proposing new rules to create a tailored offering regime for certain investment contracts involving crypto assets”), https://www.federalregister.gov/documents/2026/08/13/2026-16519/sunshine-act-meetings. ↩

18. Astraea Counsel verification, August 17, 2026, against four primary sources: the Federal Register proposed-rule record for the Commission (no crypto-offering release; one proposed rule published on or after July 1, 2026); the Federal Register notice record for the Commission from August 10, 2026 forward (no cancellation notice for the August 14 meeting; the next Sunshine Act notice, 91 Fed. Reg. 53,303 (Aug. 17, 2026), notices a closed meeting for August 20); the Commission’s Meetings & Events calendar, https://www.sec.gov/newsroom/meetings-events (listing only the August 20 closed meeting and a September 17 roundtable on 24-hour trading); and the Commission’s press releases, https://www.sec.gov/newsroom/press-releases (the August 14, 2026 release, No. 2026-75, announces an enforcement action). Trade press reported the meeting as postponed; that characterization is press-sourced and is not established by the primary record, which establishes only the absence of a release. ↩