The 600,000 SOL Ghost: Inside SkyAI's Treasury Revolt and the Quiet Death of the Reserve Narrative

The number that should stop you isn't a liquidation level and it isn't a funding rate. It's 3.50. That is the price Bastion — an activist shareholder with a lawyer's patience and a predator's timing — has pinned to SkyAI's tender offer. It is not a market quote. It is a demand, a threshold, a dare. And on the other side of that dare sits a corporate shell holding 600,000 SOL, a treasury large enough to move a mid-tier validator's delegation book, wrapped inside an entity that ninety-nine percent of Crypto Twitter cannot name without a search bar. There is no protocol upgrade in this story. No token generation event. No audit to run, no bytecode to disassemble, no sequencer to interrogate. What there is — buried under three lines of "breaking news" that most aggregators will never expand — is a governance fight that tells you more about the true state of this bull market than any airdrop of the quarter. The reserve narrative is cracking at the seams. SkyAI is where the seam splits first.

To understand why a $3.50 tender demand matters, you first have to understand what SkyAI actually is, because the category error here is the entire story. SkyAI is not a DeFi protocol. It is not a Layer 2. It does not have a TVL chart, a sequencer queue, or a governance token with an emissions schedule you can model in a spreadsheet. SkyAI appears to be a Digital Asset Treasury — a DAT — a listed corporate entity whose core balance-sheet asset is a cryptocurrency. In this case, Solana. Think MicroStrategy, but the reserve asset isn't Bitcoin. Think Metaplanet, but the flag pinned to the chart is SOL, not BTC. The company does not sell a product in the way a normal operating business does. Its "business" is holding a token and marking it to market, and its equity is, at root, a leveraged, fee-bearing, tax-dragged proxy for that token's price.

This matters because the market spent 2024 and 2025 falling in love with exactly this structure. The MicroStrategy playbook — issue equity or debt, buy the reserve asset, let the NAV premium do the compounding — was cloned dozens of times across dozens of assets. Bitcoin treasuries first. Then Ethereum treasuries. Then, inevitably, Solana treasuries, because Solana had the liquidity depth, the institutional custody rails, and a narrative hot enough to justify a premium. SkyAI is one node in that constellation. Bastion is a shareholder who has decided the constellation has stopped paying.

Here is the collision. Bastion is demanding a tender offer at 3.50 dollars per share — or per unit, the source is deliberately vague — and simultaneously demanding a restructuring of the board. And Bastion has attached a threat that should make any securities lawyer sit up: it will not participate on the current terms. That single sentence — will not participate on the current terms — converts a routine capital-return event into a hostage situation. The board wants a deal on its terms. Bastion wants a deal on its terms. In between sits 600,000 SOL and a pile of cash that covers roughly eighty percent of the offer. Nobody has published the NAV. Nobody has published the share count. Nobody has published the registration jurisdiction. And that informational vacuum is not a footnote. In an event like this, the vacuum is the trade.

Let me be forensic about the structure before I get to the punchline, because I have spent fifteen years watching crypto narratives get repriced, and I have learned that the most dangerous moment in any treasury story is the moment the story stops being about the asset and starts being about the balance sheet. Based on my audit experience — from parsing Bancor's contract in 2017 to manually unwinding the Terra rebasing mechanism in 2022 — I can tell you that the moment a structure becomes legible to activists is the moment it becomes fragile. SkyAI just became legible.

The DAT Model, Stripped of Its Marketing

A Digital Asset Treasury is a simple machine dressed in complicated clothing. It holds an asset. It issues securities. It tries to make the spread between the price of its securities and the value of its assets positive. That spread — the premium — is the entire engine. When the premium is positive, the company can issue more securities, buy more of the asset, and the per-share asset backing rises without the company doing anything except existing. It is a reflexivity machine, and like all reflexivity machines it works beautifully until it doesn't.

MicroStrategy is the archetype. For years it traded at a persistent premium to its Bitcoin holdings, which allowed it to issue convertible debt and equity at accretive prices, which allowed it to buy more Bitcoin, which fed the narrative, which sustained the premium. The loop was self-reinforcing as long as the asset trended up and the market believed the company was the best available proxy. But the loop has a hidden dependency: it requires the premium to be justified by something other than circularity. MicroStrategy's justification was scale, liquidity, index inclusion, and a founder who functioned as a human marketing department. Most DATs have none of those things. They have the structure without the moat.

SkyAI, by every signal in the source material, sits in the second category. It holds 600,000 SOL. It holds cash. It does not appear to generate operating cash flow. Its value proposition is "we are a Solana proxy you can buy in a brokerage account." That is a real proposition in a bull market, when retail and institutions alike want exposure without custody headaches. It is a worthless proposition when the premium collapses to a discount, because then the same shareholders realize they can get more SOL per dollar by buying SOL directly than by buying the wrapper. When the wrapper trades below its contents, the wrapper is a trap, and the only rational move is to force the wrapper to return the contents. That is precisely what Bastion is doing.

The smart contract never lies, and neither does a balance sheet when an activist reads it line by line. Bastion has read the balance sheet. Bastion has found the discount. Bastion has decided to extract it.

The 600,000 SOL Concentration Problem

The first technical fact that matters is not technical at all in the software sense. It is a concentration fact. SkyAI's net asset value is bound almost entirely to a single token. Six hundred thousand SOL is not a diversified treasury. It is a leveraged bet on one network's price, one network's uptime, one network's validator economics, and one network's narrative cycle. Every dollar of that NAV is a dollar of Solana beta, and every Solana beta is a dollar of correlated risk.

This is the part the marketing never says out loud. A DAT is sold to investors as "exposure." It is actually a leveraged, fee-laden, illiquid expression of exposure, and its downside is amplified by every cost the wrapper adds: management fees, listing costs, legal costs, tax drag, and the discount that appears the moment sentiment turns. When SOL rallies, the wrapper looks like genius. When SOL stalls, the wrapper bleeds in ways a direct holder never does, because the direct holder has none of the structural overhead.

Let me quantify what I can and flag what I can't, because this is where most coverage goes wrong — it either asserts numbers it doesn't have or refuses to reason at all. The source material gives me three hard anchors: 600,000 SOL in remaining holdings; cash covering approximately eighty percent of the tender; and a demanded price of 3.50 per unit. Everything else is inference, and I will label it as such.

The word "remaining" is doing enormous work in that first anchor. Remaining holdings implies prior holdings that are no longer there. A treasury that once held more SOL and now holds 600,000 has, by definition, either sold some, spent some, or had some seized or pledged. In a capital-return story, the most likely reading is that the company has already begun liquidating, or has been forced to liquidate, and that 600,000 SOL is what is left after earlier draws. That reading reframes the entire event. This is not a company deciding whether to return capital. This is a company that has already started returning capital and is now arguing with a shareholder about the price of the last tranche.

If that reading is correct, the tender is not a growth event. It is a liquidation event. And liquidation events are governed by a different logic than growth events. In a growth event, everyone wants the premium to expand. In a liquidation event, everyone wants the discount to close, and the fight is over who captures the closing spread.

The NAV Discount Engine

Here is the mechanism that turns a sleepy treasury into a battlefield. DATs, almost without exception, trade at a discount to their net asset value when sentiment cools, for four structural reasons that have nothing to do with the asset itself.

First, liquidity. A share in a small DAT is less liquid than the underlying token on a major exchange. Investors demand a discount to hold something they cannot exit instantly. Uniswap taught me liquidity is truth — the price you see is only real up to the size you can actually trade, and a thin wrapper around a deep asset inherits none of the asset's depth. Second, fees. The wrapper charges management and operating costs that the direct holder avoids. Every dollar of fee is a dollar of NAV leakage, and the market prices that leakage in advance. Third, tax and structure. A corporate holder of crypto may face different tax treatment and different reporting obligations than an individual, and those differences show up as a persistent haircut. Fourth, opacity. The less the market understands the structure, the wider the discount, because uncertainty itself has a cost. SkyAI, by the source material's own admission, discloses almost nothing — no share count, no NAV, no jurisdiction, no financial detail. That opacity is not neutral. It is a discount generator.

Put those four together and you get the classic activist setup: a company whose market price is meaningfully below the sum of its parts, run by a board that benefits from the status quo, with a shareholder who can do the arithmetic and is willing to make noise. Bastion is not a philanthropist. Bastion is an arbitrageur of structure. The 3.50 demand is not a valuation of the business. It is a valuation of the exit.

And here is the subtle point that separates a real analyst from a headline repeater: the demand for 3.50 tells you almost nothing about what SkyAI is worth, and almost everything about what Bastion thinks it can extract. A tender demand is a negotiation anchor. It is set to be high enough to be worth fighting for and low enough to be defensible in a filing. The number itself is theater. The structure around it — board restructuring plus a refusal to participate on current terms — is the actual payload.

The 3.50 Anchor and the Eighty Percent Cash Paradox

Now the arithmetic, with all caveats nailed to the mast. The source tells me cash covers roughly eighty percent of the tender. That single data point implies two things at once, and they point in opposite directions.

Implication one: the company is more cash-heavy than a pure SOL proxy would be. If eighty percent of the offer can be funded from cash, then the SOL position is not the primary funding source for the offer — it is the backstop. That suggests either that SkyAI raised a large cash cushion at some point, or that it has already monetized a substantial portion of its earlier SOL holdings into cash. Either way, the "SOL treasury" framing that presumably drove the narrative is now only partially accurate. The company is transitioning from a SOL-holding entity toward a cash-and-residual entity.

Implication two: the remaining twenty percent has to come from somewhere, and the most obvious somewhere is the SOL position. If that is right, then completing the tender requires selling SOL, and selling SOL into the market — even via over-the-counter desks — creates supply pressure. Not catastrophic supply pressure. Six hundred thousand SOL is a rounding error against a circulating supply measured in hundreds of millions. But it is not nothing, especially if it is dumped in illiquid hours or if the market reads it as a signal that a known treasury is exiting.

Let me do the back-of-envelope that the source material itself flagged as low-confidence, because even a low-confidence frame is better than no frame, as long as I label it. Suppose 600,000 SOL is worth roughly twenty percent of the total tender. If SOL trades between 150 and 250 dollars, that stake is worth somewhere between ninety million and one hundred fifty million dollars. If that is twenty percent, the total tender is somewhere between roughly four hundred fifty million and seven hundred fifty million dollars, with cash covering the first eighty percent — roughly three hundred sixty million to six hundred million dollars in cash. I want to be explicit: this is a framework, not a forecast. It rests on the assumption that SOL funds exactly the residual, which the source does not confirm. But it gives you the order of magnitude, and order of magnitude is what separates analysis from vibes.

Now the paradox. If the company has enough cash to fund eighty percent of the offer, why is there a fight at all? The answer is almost always the same in these situations: the fight is not about whether to return capital. It is about the price and the control. The board wants to return capital on terms that preserve its position and its discretion. Bastion wants to return capital on terms that maximize the spread and, crucially, replace the people setting the terms. The cash paradox resolves the moment you realize the tender is not the prize. The board seat is the prize.

The Liquidity Mechanics of Unwinding 600,000 SOL

This is where I get granular, because liquidity is where treasuries die and where narratives get tested. Six hundred thousand SOL is a real position. It is not MicroStrategy-scale, but it is large enough that how it exits matters more than whether it exits.

If SkyAI liquidates via over-the-counter desks, the market impact is minimized, but the OTC market extracts a spread, and that spread is a hidden tax on NAV. If it liquidates on exchanges, the impact depends entirely on timing and venue depth. Solana's spot depth is genuinely deep in normal conditions — it is one of the more liquid altcoin markets — but depth is not a constant. It compresses violently during volatility, and it compresses precisely when a seller is most likely to be forced to sell. This is the reflexive trap. The moment the market learns a treasury is unwinding, the price softens, which reduces the value of the remaining treasury, which may force faster unwinding, which softens the price further. It is the same shape as the dynamic that ate Terra, though at a fraction of the scale and without the algorithmic-stablecoin amplifier. Surviving the Terra algorithmic trap taught me to look for exactly this shape: a position large enough to matter to itself, in a market thin enough to be moved by its own exit.

Now layer on Solana's own structural backdrop, and I have to bring in a view I have held for a while. The post-Dencun world changed the economics of rollups and, by extension, the competitive pressure on every L1 that competes for fee revenue and developer mindshare. Blob space is a finite resource, and the market has been pricing it as if it were infinite. When that space saturates — and I expect it to saturate within roughly two years — rollup costs reset upward, and the calculus of where activity lives shifts again. Solana's value proposition has been throughput and low fees, and it has been winning on that basis. But the moment the broader scaling landscape rebalances, Solana's differentiation gets re-underwritten by the market, and treasuries that are pure Solana beta get re-underwritten with it. A DAT holding SOL is not just holding a token. It is holding a position in an ongoing argument about where blockspace economics settle. That argument is not resolved. Anyone pricing SkyAI's SOL at a static multiple is ignoring the argument entirely.

Governance: The Williams Act, Proxy Fights, and the Fight for the Board

The regulatory layer here is not a side quest. It is the most legally dense part of the entire event, and it is where the risk concentrates. Tender offers are among the most heavily regulated transactions in securities law. In the United States, the Williams Act governs them, and it imposes disclosure requirements, a minimum open period of twenty business days, pro-rata acceptance rules, and a whole architecture of anti-fraud provisions designed to prevent coercive or manipulative offers. A demand that a shareholder "will not participate on current terms" is not a casual statement. It is a signal that the terms, the process, or both are contested, and contested tenders attract lawyers the way light attracts moths.

Then there is the board restructuring demand, which points toward a different regulatory regime entirely: the proxy rules. If Bastion intends to replace directors through a proxy contest, it is operating under Regulation 14A, with its own disclosure and solicitation requirements. A proxy fight plus a contested tender is a double-barreled governance event, and each barrel has its own compliance surface. If SkyAI is a listed or reporting entity — and the existence of a "filing" in the source material strongly implies that it is — then its 600,000 SOL position carries accounting and disclosure obligations that the market may be undervaluing. How do you mark a 600,000 SOL treasury? At fair value, presumably, which means every SOL price swing flows through the financials, which means the reported numbers are volatile by construction. That volatility is not a bug the board can hide. It is a feature the filings must expose.

I do not have the jurisdiction, the listing status, or the legal structure, and I will not pretend to. What I can say is that the combination — micro-cap or small-cap entity, crypto reserve on the balance sheet, activist shareholder demanding a board overhaul — is one of the most litigation-prone configurations in public markets. If there is a flaw in the tender process, it will be found. If there is a flaw in the disclosure, it will be found. The absence of information in the source material is not evidence that the process is clean. It is evidence that we are looking at it through a keyhole.

What the Board Is Actually Afraid Of

Here is the part no aggregator will write, because it requires reasoning past the headline. When a board resists a tender at a higher price, the naive read is that the board is protecting shareholders from an undervalued offer. The cynical and, in my experience, more accurate read is that the board is protecting its own discretion. A board that controls a 600,000 SOL treasury controls the timing of every sale, the choice of every counterparty, the selection of every custodian, and the discretion to spend treasury assets on operations, compensation, and strategy. A tender that returns capital to shareholders strips that discretion. A board restructuring that replaces the directors strips it faster. Bastion is not just asking for money. Bastion is asking for the keys.

And when the keys are on the table, the valuation question becomes a question about whose assumptions win. If the board values SOL conservatively and Bastion values it optimistically, then the fair tender price is genuinely disputed, and the dispute is not bad faith — it is a real disagreement about the future of an asset. If the board values SOL conservatively and still refuses to tender at a price above its own mark, then the dispute is about control, not valuation. I cannot tell you which is true from the source material. But I can tell you that the market will eventually tell us, because the filings will force the assumptions into the open.

This is also where the DAT model reveals its deepest flaw. A DAT with no operating business has no organic way to resolve disputes. A normal company can point to product revenue, growth, a pipeline. A DAT can only point to its holdings, and its holdings are public, and everyone can do the same arithmetic. When the only asset is public and the only liability is a fee stream, the discount becomes a permanent negotiation rather than a temporary inefficiency. Bastion is not an anomaly. Bastion is the logical endpoint of the model.

The Solana Supply Question, Sized Honestly

Let me size the SOL impact honestly, because honesty is where credibility lives. Six hundred thousand SOL, against a circulating supply in the hundreds of millions, is a fraction of a percent. As a standalone supply event, it is negligible. Any article that tells you SkyAI's potential liquidation will "crash Solana" is writing fiction. The real effects are narrower and more interesting.

First, venue effects. If the liquidation routes through exchanges in concentrated size, it can dent local order books in low-liquidity windows, producing wicks that look dramatic on a one-minute chart and mean nothing on a weekly. Second, signal effects. A known treasury exiting is a narrative event even when it is not a supply event. Other DAT holders watch other DAT holders. If one Solana treasury unwinds, the market re-underwrites the others, and the entire cohort can reprice toward NAV simultaneously. That cohort-wide repricing is the real risk, and it is a narrative risk, not a liquidity risk. Third, validator effects. If SkyAI had staked any portion of its SOL, its exit would reduce delegated stake somewhere. The source does not say whether the SOL is staked, and that omission matters. An unstaked treasury is a dormant position. A staked treasury is an active participant whose exit reshuffles a validator's economics. I cannot resolve this from the source, so I flag it as an open question rather than pretending to an answer.

The honest conclusion: SkyAI's SOL is a small supply event and a large signal event. Price the signal, not the supply.

The Narrative Cycle Nobody Wants to Name

The deepest thing about this story is not the numbers. It is the timing. Activist campaigns are, almost by definition, late-cycle phenomena. They appear when the appreciation story has stalled and the value-realization story takes over. When shareholders stop asking "how much higher can this go" and start asking "how do we get our money back," the narrative has already peaked. Bastion's arrival is not a bullish catalyst for the DAT theme. It is a symptom that the theme has matured into a distribution problem.

The source material frames this as a possible industry inflection: if SkyAI's event ends in capital return, it could weaken the narrative premium of other DATs and pull the whole cohort toward NAV. I think that framing is directionally right, and I would push it further. The DAT model worked in 2024 and 2025 because the market believed the reserve assets would keep appreciating and because the premium gave issuers a self-funding engine. Both conditions are cyclical. The premium is cyclical because sentiment is cyclical. The appreciation is cyclical because everything is cyclical. A structure that depends on a positive premium and a rising asset to function is a structure that inverts violently when either condition fails, and activists are the mechanism of that inversion.

Chasing alpha through the 2017 hallucination taught me to be suspicious of any structure whose value proposition is "buy our wrapper instead of the thing." In 2017 that wrapper was a token sale with a whitepaper. In 2025 and 2026, the wrapper is a corporate shell with a treasury. The shape is identical. The pitch is "we give you exposure, plus upside, plus convenience." The reality is "we give you fees, plus opacity, plus a discount you will eventually be forced to eat." When the wrapper trades below its contents, the wrapper has failed, and the market's response — force the wrapper to return the contents — is not a failure of the market. It is the market working.

The Contrarian Angle: The Revolt Is the Top Signal

Now the counter-intuitive part, the part I have been building toward and the part the consensus will miss. Everyone will read this as a story about value being unlocked. Bastion demands a higher tender, the board resists, and if Bastion wins, shareholders get paid. Bullish. Right? Wrong. The bullish framing is the trap. Read the sequence of events as a cycle, not as a single transaction.

When a reserve-accumulation narrative is young, insiders and believers accumulate, the premium expands, and the wrapper sells itself. When the narrative is mature, the premium compresses, the discount appears, and insiders quietly look for exits. When the narrative is late, outside activists arrive and demand the exit on behalf of everyone. The arrival of the activist is not the beginning of the value-realization phase. It is the announcement that the accumulation phase is over and that the smart money has already decided the future is distribution, not appreciation. Bastion is not a growth investor. Bastion is a liquidator with a legal team. The presence of a liquidator is the market's way of saying the upside case has been quietly abandoned by the people closest to the numbers.

This is the inversion that the bull-market framing cannot see. A bull market trains you to read every governance event as a value-unlock opportunity, because in a bull market, value unlocks are bullish. But value unlocks are also, structurally, exits. When the thing being unlocked is a treasury that was supposed to appreciate forever, the unlock is the moment the forever ends. The correct read of SkyAI is not "shareholders might get 3.50." The correct read is "a Solana treasury has decided that holding Solana is no longer the best use of its capital, and it took an activist to force the admission into the open." That is a much darker and much more informative statement.

There is a second contrarian layer. The consensus assumes the tender will resolve cleanly — either the board caves and pays, or Bastion walks and the status quo holds. The consensus is underweighting the third outcome: a prolonged governance stalemate with legal costs, disclosure fights, and a paralyzed treasury that neither returns capital nor deploys it. In a stalemate, the SOL sits, the fees accrue, the discount widens, and everyone loses except the lawyers. Stalemate is the most expensive outcome and the least discussed, and it is the one I would assign meaningful probability to, because the structural conditions for it are all present: a board with discretion to protect, an activist with a mandate to extract, and an information vacuum wide enough for both sides to litigate indefinitely.

Curating chaos for clarity means naming the outcome nobody is pricing. The outcome nobody is pricing is gridlock.

What This Says About the Broader Reserve Trade

Step back and look at the pattern across the entire DAT landscape. Bitcoin treasuries. Ethereum treasuries. Solana treasuries. Each one was sold on the same logic: accumulate the asset, let the premium compound, become the institutional proxy. Each one faces the same terminal condition: when the premium inverts to a discount, the structure becomes a value-extraction problem, and activists or liquidators arrive to solve it. SkyAI is not unique. SkyAI is early. The Solana cohort is smaller and less liquid than the Bitcoin cohort, so it will crack first, and SkyAI is where the crack is visible.

I should be careful to distinguish the MicroStrategy class from the SkyAI class, because conflating them is the most common analytical error. MicroStrategy has scale, index inclusion, deep debt markets, and a founder who is himself the marketing department. It can roll its structure for years because its premium is defended by real advantages. SkyAI, by the source material's own signals, has none of those defenses. Its only defense is the board's discretion, and that defense is exactly what Bastion is attacking. The DAT model is not uniformly fragile. It is fragile in proportion to the absence of a moat, and SkyAI's moat is a rumor.

This is also where the Bitcoin-versus-altcoin reserve distinction matters, and I want to make a point I have made before in a different context. Ordinals injected a new narrative and a new fee-revenue stream into Bitcoin at a moment when the security-budget debate was genuinely uncomfortable. Without that inscription wave, Bitcoin's long-term security model would already be under harder scrutiny. The lesson is not that inscriptions are good. The lesson is that even Bitcoin — the deepest, most liquid, most institutionally embraced reserve asset on earth — needed a fresh narrative to keep its economics defensible. If Bitcoin needed that, a Solana treasury wrapper with a fraction of the liquidity and none of the institutional embrace needs it ten times more. SkyAI did not get a fresh narrative. SkyAI got an activist.

The Question of Price Discovery

One more analytical thread before I close, because it is the thread that determines whether anyone makes money here. The entire event hinges on price discovery, and price discovery is broken in this story. There is no published NAV. There is no published share count. There is no published jurisdiction. There is no published cash balance in absolute terms. Every actor in the market is trading against an estimate, and estimates are where edges live and where losses hide. Bastion has an estimate. The board has an estimate. The market has an estimate. If the estimates diverge — and in an information vacuum they always diverge — then the tender is a dispute about whose estimate becomes the settlement price.

If the discount is real and large, there is a genuine arbitrage for anyone who can assemble the numbers: buy the wrapper below NAV, push for the return of the contents, capture the closing spread. That is Bastion's trade, and it is a legitimate one. But the arbitrage is only as safe as the disclosure is complete, and the disclosure here is not complete. An arbitrage built on an estimate is not an arbitrage. It is a bet with an arbitrage-shaped costume. I would want the filing, in full, before I treated the discount as a fact rather than a hypothesis.

And this is the final structural point. The opacity that created the discount is the same opacity that makes the discount dangerous to capture. The board benefits from opacity because opacity protects discretion. Bastion benefits from opacity because opacity makes the discount look bigger than it might be. Both sides have an incentive to keep the numbers fuzzy for as long as possible. The loser in a fuzzy number is always the public shareholder, who cannot price the risk and cannot time the exit. The tender, whatever its outcome, will eventually force some clarity into the open, and that clarity is the only thing in this story that is unambiguously good.

Takeaway: What to Watch, and Why It Matters Beyond SkyAI

Watch three signals and ignore the noise. First, the tender document itself: does the price move toward 3.50, does the board publish a NAV, and does the filing reveal the jurisdiction and the share count. That single document will resolve most of the fog. Second, the chain: if 600,000 SOL begins moving to exchange addresses or OTC desks, the liquidation hypothesis is confirmed, and the signal matters more than the supply. Third, the cohort: if other Solana and altcoin treasuries start drawing activists, SkyAI is not a story — it is the first chapter of a repricing across the entire reserve trade.

The reserve narrative did not die with a crash. It will die with a tender offer, a board fight, and a shareholder who did the arithmetic. The question worth carrying forward is not whether Bastion gets its 3.50. It is whether the market has finally understood that a wrapper around an asset is not the asset — and that when the wrapper cracks, the crack always starts where the treasury is largest and the transparency is thinnest. SkyAI is that place. Watch what crawls out of the crack.