Reading the Silence Between the SPVs
On the morning of October 9, 2026, the Securities and Exchange Commission's San Francisco Regional Office filed a civil complaint against two former Linqto executives โ Sarris and Endoso โ and the filing named the men first, the company second. That ordering is not clerical. In my decade of watching enforcement paper cross the wire, the sequence tells you everything: when the Commission lists individuals before the entity, it has already concluded the entity is a husk and the recoverable assets live in two personal balance sheets. The charges themselves read like a compliance textbook turned inside out โ unregistered securities under Section 5, fraud under Section 17(a) and Rule 10b-5, and, buried three pages in, a claim that almost nobody in the retail-facing coverage noticed: that Linqto operated as an unregistered investment company. Three independent theories of liability, stacked like backup charges in a criminal indictment, and the one that will actually decide the case is the one the least people are talking about. I want to walk you through the architecture, because the story here is not a rogue salesman. The story is a structure โ a deliberately fragmented one โ and structures, unlike people, do not lie or tell the truth. They simply do what they were built to do.
Context: What a Pre-IPO SPV Actually Is
To understand why this complaint is a structural verdict rather than a personality verdict, you have to understand the instrument at the center of it. A special purpose vehicle, in the pre-IPO context, is a legal container. A platform like Linqto wants to sell retail investors a slice of a private unicorn โ a company that has not yet gone public, whose shares are not traded on any exchange. But the unicorn does not want ten thousand small shareholders on its cap table. It wants a handful of large, quiet, institutional holders. So the platform forms an SPV, the SPV buys a block of shares from an existing holder or from the company in a tender, and then the platform sells interests in the SPV to retail buyers. The retail buyer never touches the unicorn's cap table. They hold a fractional interest in a wrapper that holds the shares.
The legal machinery here is not exotic; it is well-trodden. Under the Howey test โ the four-prong standard the Supreme Court laid down in 1946 and has never abandoned โ an interest in that SPV is an investment contract. Money is invested, in a common enterprise, with an expectation of profit, derived from the efforts of others. Every prong is satisfied by design. That makes the SPV interest a security, and the moment it is a security, Section 5 of the Securities Act of 1933 snaps into place: you cannot offer or sell it to the public without registration or a valid exemption.
The exemption everyone reaches for is Regulation D. Rule 506(b) and 506(c) allow a private issuer to sell securities without registration, provided the buyers are accredited investors โ a net-worth and income threshold that excludes the overwhelming majority of retail participants. Rule 506(c) even permits general solicitation, but only if the issuer takes reasonable steps to verify that every single buyer is accredited. That verification requirement is the load-bearing wall of the whole structure. Remove it, and the entire building falls into Section 5's strict-liability basement.
And there is a second container nested inside the first. The Investment Company Act of 1940 regulates entities that are primarily in the business of investing in securities โ which is precisely what an SPV is. Section 3(c)(1) provides an exemption from registration, but it caps the fund at one hundred beneficial owners, and those owners must themselves be qualified purchasers or accredited investors. Section 3(c)(7) raises the ceiling but requires every holder to be a qualified purchaser, a much higher bar. The exemption is not a loophole; it is a narrow gate with a guard posted at it. The question the SEC's complaint poses is whether Linqto walked through that gate or dismantled it.
So that is the context. A legal container designed to let a small number of sophisticated buyers hold a private asset, wrapped in an exemption designed to keep unsophisticated buyers out, and sold by a platform whose entire commercial premise is to let those unsophisticated buyers in. The tension is not incidental to the model. The tension is the model. And when a business is built on a tension, the only question is how long the regulator lets it breathe.
Core: The Two Independent Paths the SEC Picked
Here is where the complaint becomes interesting, and where the retail-facing coverage largely failed. Most reporting focused on the fraud counts โ the false statements about pricing, about sellouts, about an algorithm that supposedly set share prices but was, according to the Commission, a human being with a spreadsheet. Those are serious. They carry scienter requirements, which means the SEC must prove the defendants knew the statements were false. That is a higher bar. But the complaint does not rest on them alone, and it does not need to.
The Commission pleaded two structurally independent violations that require no proof of intent whatsoever. The first is Section 5 โ unregistered sale of securities. Section 5 is a strict-liability statute. It does not ask whether you meant well. It does not ask whether the buyers understood the risk. It asks one question: was the security registered or validly exempt? If the answer is no, liability attaches. Full stop. I have watched defendants walk into Section 5 proceedings with flawless intentions and perfect disclosure and walk out with a judgment against them, because intent is simply not an element. When the SEC stacks a strict-liability count next to a fraud count, it is building a floor beneath its own case. Even if the fraud theory collapses โ even if a jury finds the pricing statements were merely sloppy rather than knowingly false โ the Section 5 count stands on its own.
The second path is the one that deserves a full forensic reading: the claim that Linqto operated as an unregistered investment company under Section 7 of the 1940 Act. This is not a technicality tacked onto the end of the complaint. It is the structural fingerprint of the business. The Commission is effectively alleging that Linqto did not simply sell securities improperly โ it ran an investment company without the registration that would have subjected it to disclosure, leverage limits, custody rules, and independent governance. When a regulator pleads both "you sold unregistered securities" and "you ran an unregistered fund," it is telling you it believes the entity was operating a parallel, unregulated asset-management business under a retail-facing brand.
The Parallel SPV Hypothesis
Here is the part I find most forensically revealing, and it is the part almost nobody has connected. If Linqto was running an investment company, how did it avoid the one-hundred-beneficial-owner cap in Section 3(c)(1)? The answer, I suspect, is the same answer that shows up in every structural evasion I have audited: parallel SPVs. You do not put ten thousand investors into one fund. You create dozens, or hundreds, of separate SPVs โ one per asset, or one per tranche, or one per sales channel โ each holding fewer than one hundred beneficial owners, each claiming its own Section 3(c)(1) exemption. On paper, no single entity breaches the ceiling. In substance, the same operator, selling to the same retail population, through the same interface, has assembled a thousand-investor exposure by slicing it into legally distinct pieces.
I have seen this pattern before. In 2020, during the yield-farming frenzy, I spent weeks stress-testing the emission schedules of forks that claimed to be distinct protocols while sharing liquidity pools, governance keys, and deployer wallets. The structural lesson from that period transfers directly: when a single economic enterprise is partitioned into nominally independent legal or technical containers, the partitions are almost never about investor protection. They are about the perimeter of the regulator. The fragmentation is the product. The cap is not respected; it is routed around.
If that is what happened at Linqto, then the accredited-investor verification requirement under Rule 506 was likely degraded in parallel. You cannot verify accreditation across a population you have deliberately structured yourself not to see as a single population. The fragmentation that defeated Section 3(c)(1) also defeated Rule 506(c)'s verification mandate, because the platform could plausibly claim that each SPV had "done its best" while the aggregate operation sold to anyone with a checking account. Two independent regulatory limits, defeated by a single structural decision. That is not sloppiness. That is design, and design is what the SEC prosecutes when it wants to establish precedent rather than merely punish a person.
The $430 Million Disgorgement Question
Now follow the money, because this is where the 2021 National Defense Authorization Act quietly changed the arithmetic of every fraud case that followed it. Before that statute, the SEC's power to claw back ill-gotten gains was a creature of case law, and Kokesh v. SEC โ decided in 2017 โ had classified disgorgement as a penalty. That classification dragged disgorgement under the five-year statute of limitations that governs penalties. Five years. That was the ceiling on how far back the Commission could reach.
Section 6501 of the 2021 NDAA wrote disgorgement directly into the Securities Exchange Act, at Sections 21(d)(3)(A)(ii) and 21(d)(7), and โ critically โ established a ten-year limitations period for it. This is not a footnote. It is the difference between recovering a slice of the wrongdoing and recovering the whole thing. If Linqto's sales ran continuously from 2021 through 2024, as the complaint's timeline suggests, then the entire window sits inside the ten-year reach. The potential disgorgement base is the full scale of the retail sales โ a figure the complaint places in the neighborhood of 430 million dollars โ not the fraction that would have been recoverable under the old five-year rule. When I read enforcement filings, I always trace the limitations arithmetic first, because it tells me what the Commission actually thinks it can collect, and that tells me how hard it intends to fight.
And then there is Liu v. SEC, decided in 2020, which cut the other way. Liu held that disgorgement cannot exceed the wrongdoer's net profits and must be awarded for the benefit of investors. For the defendants, this is the single most valuable defensive tool they have. It opens the door to arguing that legitimate operating costs, and any funds already returned to investors, should be deducted from the disgorgement base. A defense built on Liu could shrink the exposure meaningfully. But Liu has a limit, and the SEC will aim straight at it: if the SPV cash flows were cyclical or self-referential โ if new investor money was used to pay obligations owed to earlier investors or to the platform itself in a pattern resembling a Ponzi structure โ then the net-profit logic collapses, because there were no legitimate profits to net against. Whether the cash flows were a functioning business or a rotational scheme is the factual hinge on which the entire damages phase turns.
The Criminal Overlay
I want to spend real time on the criminal dimension, because the civil complaint is only half the story, and the quieter half is the more dangerous one. The Commission did not investigate this alone. The complaint credits the Southern District of New York and the FBI. When the SEC's regional office brings in SDNY federal prosecutors and the Bureau, it is not asking for help with document review. It is signaling that the conduct may be criminal, and that the Commission is coordinating with โ or effectively feeding โ a parallel criminal track.
If a criminal case follows, the relevant statutes are 18 U.S.C. ยง1348, the securities fraud statute, and ยง1343, wire fraud. The maximum penalties are twenty-five years and twenty years respectively. Read that again, because it reframes the entire risk profile. The civil exposure is money โ disgorgement, penalties, an officer-and-director bar. The criminal exposure is liberty. No amount of Liu netting or Section 5 ambiguity protects against a criminal conviction, and the standard of proof that matters there is beyond a reasonable doubt, which cuts both ways: harder for the government to secure, but catastrophic if it lands.
The presence of SDNY also changes the discovery dynamics. Federal prosecutors have subpoena power, grand jury tools, and the ability to pressure cooperators. In a two-defendant case with a fragmented corporate structure, the first person to cooperate typically defines the case against the second. I have watched this dynamic from the reporting side during the Terra/Luna collapse in 2022, when I interviewed four former associates of the founder and watched how the narrative of "decentralized stability" was dismantled not by journalists but by insiders who understood the mechanism from the inside. In a criminal SPV case, the same force applies with more force: the structure was built by specific people who kept records, and those records become leverage.
Where the Evidence Actually Lives
Let me get forensic about the "algorithm" claim, because it is the most under-analyzed piece of the complaint and, in my judgment, the most dangerous one for the defendants. The SEC alleges that Linqto represented that share prices were set by an algorithm, when in fact they were set manually. This is what I would call a hard fraud allegation, as opposed to a soft one. A soft fraud is a rosy projection, an optimistic valuation, a forward-looking statement that turned out wrong โ all of which carry built-in scienter defenses, because everyone knows projections are uncertain. A hard fraud is a factual claim about a mechanism that either exists or does not. Either there is code that computes prices or there is not. Either there is a system that clears orders or there is a person doing it by hand.
A factual claim about a mechanism is falsifiable by the mechanism's own exhaust. System logs. Version histories. Deployment records. Access timestamps. If a pricing engine existed, it left a trail: commits, config files, a runtime, a monitoring dashboard. If it did not exist, its absence is equally legible โ the records show a person typing numbers into a spreadsheet, and the spreadsheet's metadata shows when and by whom. This is why I keep returning to the same principle that has organized my work since 2017, when I dissected the ERC-20 and multisig contracts that the market had pronounced "safe" and found three reentrancy vulnerabilities the hype had buried. The audit trail never lies. A narrative can be curated. A ledger cannot be un-committed. When a defendant claims a mechanism, the mechanism's logs are the witness, and logs do not take the Fifth.
This is also why the internal electronic evidence is likely to drive a settlement rather than a trial. When a fraud allegation can be confirmed or refuted by the defendant's own infrastructure, the defense loses the ability to control the factual narrative. You cannot argue to a jury that the pricing was algorithmic when the prosecution holds a commit history showing it was manual. The hard-fraud count, if supported by internal records, does not merely add to the case. It forecloses the settlement posture that would otherwise let the defendants negotiate down.
The Investment-Company Count Is the Real Precedent
Here is where I will part ways with the consensus reading. The market is treating this as a fraud case with a regulatory garnish. I think it is the reverse. The fraud counts are the part that makes headlines and drives the Liu analysis. But the unregistered-investment-company count is the part that will reshape the industry, because it attacks the structure rather than the salesmanship. If the Commission successfully establishes that a platform running dozens of parallel SPVs to sell retail access to private unicorns was operating an unregistered investment company, it has effectively declared the architecture itself non-compliant โ not this platform's version of it, but the template every platform in the category uses.
Consider what that means for the field. The pre-IPO retail-access model has always lived in a specific gap: between Regulation D's private-placement exemption and the Investment Company Act's registration regime. The pitch was that each SPV was a discrete private placement, and the platform was merely a marketplace connecting sophisticated sellers with accredited buyers. But the economics of the business required volume โ enough buyers to make the spread worthwhile โ and volume is exactly what the hundred-owner cap forbids. The only way to have both the private-placement posture and the volume economics is to fragment, and fragmentation is precisely what the investment-company count reaches. The SEC is not just saying these two men lied. It is saying the container they built to sell a lie was itself unlawful, independent of the lie.
This connects to something I have argued for years about a different corner of the market. The RWA โ real-world asset โ tokenization narrative has spent three years promising that traditional finance would migrate onto public rails. It has not, and it will not, for a structural reason: institutions do not need a permissionless public chain to move a treasury bond or a private equity stake. They need a compliant, permissioned, auditable container that their counsel will sign off on. Linqto is the retail-facing mirror of that same delusion. The premise was that you could take a private-market instrument, wrap it in a structure that satisfied the exemptions on paper, and distribute it to the public. The investment-company count says the wrapper was never exempt. The container is the compliance, or the container is the crime. There is no version where the container is neutral.
The Layer-2 Analogy Nobody Is Making
I want to push the structural reading further, because there is a pattern here that crosses asset classes, and I have been tracking it since the Layer-2 wars. There are now dozens of Layer-2 networks, each with its own bridge, its own sequencer, its own token, its own liquidity incentive program โ and they are all fighting over the same finite pool of users and capital. The industry calls this scaling. It is not scaling. It is slicing already-scarce liquidity into fragments and calling the fragmentation progress. Each L2 is nominally independent, nominally sovereign, and collectively they have fragmented the very user base that was supposed to justify them.
The Linqto SPV structure is the same pattern applied to securities law. Each SPV is nominally independent, nominally exempt, nominally under the hundred-owner cap. Collectively, they fragmented a single retail population into legally distinct pieces so that no piece breached a limit designed to protect that population. The fragmentation is not a feature of scaling. It is a technique for evading a perimeter, and it works identically whether the perimeter is a hundred-owner cap or a shared liquidity pool. When I look at the Layer-2 map, I see dozens of compliant-looking containers doing to liquidity what the parallel SPVs allegedly did to investor protection. The regulatory pressure that just landed on Linqto is, structurally, the same pressure that will eventually land on the fragmentation of L2 liquidity โ not because fragmentation is illegal, but because fragmentation in service of evading a protective limit eventually gets noticed.
Contrarian: The Real Winner Is Not the SEC
Here is the contrarian read, and I will state it plainly because the data supports it even if the sentiment does not. The biggest commercial beneficiary of the Linqto enforcement action is not the SEC, and it is not retail investors. It is the traditional licensed broker-dealer.
Run the logic. If the investment-company count establishes that fragmented SPV platforms cannot lawfully sell retail access to pre-IPO equity, then the demand for that access does not disappear. Retail appetite for private-market exposure is real and durable. What disappears is the supply channel that could serve it without a license. The demand then flows to whichever entity can serve it with a license โ and that entity is a registered broker-dealer, a firm that already has FINRA membership, net-capital compliance, supervisory procedures, and the institutional relationships to source the assets. The enforcement action does not kill the business of retail pre-IPO access. It transfers that business from the unlicensed to the licensed. The platforms that built their model on avoiding broker registration lose; the incumbent firms that already pay for registration win.
This is the part the industry's reflexive anti-regulation posture keeps missing. The category was never going to be abolished. It was going to be reallocated. And the reallocation runs toward the players who treat compliance as a moat rather than a cost. I have made this argument in a different register since the spot Bitcoin ETF approvals in January 2024, when I analyzed the flow data from the major institutional products and argued that the ETF wrapper would tame Bitcoin's volatility while increasing its correlation to equities โ a contrarian view that annoyed the "digital gold" faithful but tracked the data. The pattern is the same: a regulatory container does not destroy the underlying demand. It captures it, concentrates it, and hands it to whoever is already inside the perimeter. The pre-IPO access market is about to learn what the Bitcoin market learned two years ago.
There is a second contrarian point, and it is about who pays. The consensus assumes that enforcement protects retail. Often it does. But in a case with a 430-million-dollar disgorgement base, a parallel criminal track, and a corporate entity that may already be insolvent, the realistic recovery for the defrauded investors is a fraction of their losses, arriving years late, after the SEC's own claims and the criminal forfeiture are satisfied. Meanwhile the compliance cost of the transformation โ the licensing, the verification systems, the legal and audit infrastructure โ gets priced into the next generation of retail access, making it more expensive for exactly the population the enforcement was meant to protect. Investor protection is real, but it is not free, and the invoice does not go to the platforms. It goes to the next retail buyer.
The Governance Signal in the Officer-and-Director Bar
One more structural reading, and then the forward view. The SEC is seeking an officer-and-director bar against both men. Of all the non-monetary remedies available to the Commission, this is the heaviest. It is not a fine; it is a profession exclusion โ a determination that these individuals should not be permitted to sit on the boards or in the executive suites of public companies, potentially for years. The Commission reserves this relief for defendants it assesses as presenting a risk of recurrence and a danger to the industry, not merely a one-time lapse in judgment.
What that tells me is that the Commission's theory of the case is not "these men made a mistake." It is "these men are the kind of operators the industry needs protection from." That framing matters because it signals how the agency will treat the category in the months ahead. When the SEC seeks a bar, it is building a record for the next case. An officer-and-director bar in a case like this is not the end of an investigation. It is the opening exhibit of a sweep.
I have watched this dynamic before, in 2022, when I interviewed former associates of the Terra founder and analyzed the peg mechanism under stress rather than joining the blame game. The lesson from that reporting was that the collapse of a narrative precedes the collapse of the mechanism, and the regulator's job is to reconstruct the narrative timeline from the inside. Here, the narrative was "democratized access to private markets." The mechanism was a set of parallel SPVs and a phantom algorithm. The regulator is now reconstructing the timeline, and the officer-and-director bar is the tell that it intends to use the reconstruction as a template.
Takeaway: The Sweep Is the Story
The question is not whether Sarris and Endoso are liable. On the Section 5 count, strict liability makes liability nearly certain, and the hard-fraud allegation about the algorithm, if supported by internal records, makes settlement the rational path. The question that matters for the next eighteen months is whether the investment-company theory survives contact with the court โ because if it does, the SEC will not stop at Linqto. It will run a sweep across the entire category of retail-facing private-market platforms, using the same two-count template: unregistered securities under Section 5, unregistered investment company under Section 7 of the 1940 Act, with the fraud counts as the pressure that forces early resolution. The parallel-SPV structure that built this category is the same structure that will dismantle it.
So watch three things. Watch whether the investment-company count survives a motion to dismiss, because that is the precedent that matters. Watch whether SDNY returns an indictment, because a criminal filing converts a compliance problem into a liberty problem and accelerates every settlement in the category. And watch where the retail demand goes when the unlicensed channel closes โ because it will not evaporate. It will flow to the licensed broker-dealers, and the retail buyer will pay a compliance premium for the same exposure they used to get for free. The container was always the product. The only open question is who gets to hold it โ and at what price. When the enforcement action ends and the disgorgement is collected and the bars are imposed, the private-market access machine will still be running. It will just be running under a different owner, with a license on the wall and a markup on the ticket. The logic gates behind the yield were always set by whoever controlled the container. The only thing the SEC just changed is who is allowed to control it.