The parsed data is one sentence buried in a press release. Eligible retail and institutional users can deploy tokenized shares of Nvidia, Apple, Tesla, and three other US companies across Bybit's trading and lending rails. Six equities. Two products. One claim embedded in the copy: that blockchain tokens can cleanly represent beneficial ownership of American blue-chip stocks.
Cleanliness is the last thing I expect from this structure.
I have traced asset custody chains across this industry for eight years. I have audited blockchain projects where "decentralized" meant one multisig wallet on a cloud server. I have modeled liquidation cascades for lending protocols whose collateral factors were calibrated to a rising market and failed in a falling one. The failure was never in the token. It was always in the chain between the token and the asset it claims to represent.
The data shows tokenized equities are not an innovation. They are the third iteration of a promise that has collapsed twice before. The collapse mechanism is never the blockchain itself. It is the bridge between the on-chain representation and the off-chain reality. That bridge has already cost this industry approximately $2.5 billion in cross-chain failures, and the structural pattern is identical: custody, validation, and timing gaps at the boundary of the chain.
Tracing the ledger back to the zero-day exploit starts with one question: who holds the physical shares? Everything else is derivative of that answer. And in Bybit's announcement, the answer remains deliberately vague.
Context: The RWA Land Rush and Bybit's Positioning
Let me establish what this actually is.
Bybit ranks consistently among the top five global exchanges by derivatives volume. Over the past two years, the exchange has expanded beyond its core perpetuals business: spot trading, an emerging DeFi integration layer, and now real-world asset products. Tokenized equities are the newest frontier in that RWA push.
The mechanics are straightforward on the surface. A regulated issuer acquires actual shares of Nvidia through a licensed broker-dealer. Those shares are transferred to a custodian, typically within a special purpose vehicle, or SPV. The SPV becomes the legal owner of the shares. Tokens are then minted on a blockchain — Ethereum in most cases — representing proportional claims on the SPV's holdings. Users trade these tokens on Bybit. Users lend these tokens through Bybit's lending book.
Each step in that chain must be verified independently. My experience is that most marketing materials do not properly distinguish between what is verified, what is assumed, and what was never checked at all.
The RWA sector has become one of the few narratives attracting institutional capital in this bear market. BlackRock's BUIDL fund crossed half a billion dollars in early assessments. Ondo Finance, Backed Finance, Securitize, and a host of regulated issuers are competing to tokenize everything from money market funds to private credit to equities. The total addressable market rhetoric is measured in trillions.
Bybit is late to this party, but it is arriving with scale. The exchange's distribution muscle is not trivial. If Bybit were merely acting as a distribution partner for a compliant issuer, the margin for error would be narrower. If Bybit is creating its own standalone product, the complexity of its obligations rises significantly.
The difference matters. And the announcement does not clarify which structure applies.
I should flag the market context explicitly. This is a bear market. Capital preservation matters more than upside potential. My readership has shifted from asking "how do I make money" to asking "is my capital safe." A tokenized equity product, whatever its merits, must survive volatility in the underlying stock, volatility in the digital asset market, and operational risk from the issuer-custodian chain. That combination is untested at scale.
Let me also address the company selection. Nvidia, Apple, and Tesla are named. The three unnamed companies are almost certainly Alphabet, Amazon, and Microsoft. This is the standard mega-cap basket. The market expectation in 2025 was that tokenization would expand beyond mega-caps into small caps, emerging markets, and private markets. A curated list of six glorified blue chips undermines the access narrative that RWA tokenization claims. It also tells us something about the issuer's legal constraints: high-market-cap, highly liquid US stocks are the easiest assets to tokenize under most regulatory frameworks. The product's selection criterion is legal feasibility, not user demand.
Core Analysis: A Systematic Teardown
1. Custody Chain Anatomy
Let me perform a layer-by-layer breakdown of who holds what, and where the claims actually sit.
Layer one is the underlying asset. The issuer must buy actual Nvidia shares through a regulated broker. This is not theoretical — there must be trade confirmations, settlement records, and account statements. The purchase happens through standard clearing pathways: DTCC for US securities, or local equivalents in other jurisdictions.
Layer two is the custodian. Once purchased, the shares must sit somewhere. The key question: segregated client account or general omnibus account? Segregation is the difference between a claim that can be traced to specific assets and a claim against a pooled reserve. I have seen custody structures where tokenized assets were backed by a pool of shares that also supported other products. This is rehypothecation risk, and it is the silent killer of asset-backed tokens.
Layer three is the SPV. The SPV is a bankruptcy-remote entity designed to hold assets for token holders. The critical issue is whether the SPV is genuinely bankruptcy-remote. If the issuer goes bankrupt, can its creditors attach the SPV's assets? If the SPV is consolidated into the issuer's balance sheet, the entire token mechanism collapses. Legal opinions on bankruptcy remoteness are expensive to obtain and rarely published in marketing materials.
Layer four is the blockchain. Tokens exist on a public ledger. The smart contract that mints and burns tokens is the only on-chain component. This is the part I can actually audit directly.
Audit the code, ignore the cult. The marketing does not matter. What matters is the mint-burn authorization logic. Can new tokens be issued without new shares? If the contract has a backdoor mint function accessible to the issuer's ops team, the entire collateral model is invalid. If the burn function requires manual approval rather than being triggered automatically upon share delivery, then redemption arbitrage is delayed, and the token can trade at a persistent discount.
Based on my 2025 RWA audit for a Gulf-based financial institution, I spent six weeks examining smart contract interactions between a token minting contract and traditional banking APIs. I identified two critical vulnerabilities in the oracle data feed process. Those vulnerabilities were not theoretical. They were exploitable with moderate technical skill and would have allowed a bad actor to inflate the value of tokenized collateral by manipulating the on-chain price feed.
The takeaway from that engagement, which prevented an estimated $10 million loss: the custody chain is only as strong as its weakest layer. The weakest layer in most RWA products is never the on-chain contract. It is the set of off-chain dependencies and the oracle infrastructure that connect them.
2. What You Actually Own: The Legal Structure
The legal reality of tokenized equities is messier than the marketing suggests.
When you buy a tokenized Nvidia share through Bybit, you do not own a share of Nvidia. You own a contractual claim against the SPV. The SPV owns the actual shares. If you want to exercise shareholder rights — voting at the AGM, receiving dividends, participating in rights issues — you must go through the issuer and the custodian.
In practice, this means you hold a derivative instrument. It is economically similar to a share but legally distinct. The question is whether the derivative is properly registered and supervised.
Here is the structural problem. Most tokenized equity issuers operate under a specific securities regime. Backed Finance operates under Swiss securities law. Ondo Finance structures some products under US securities rules. The regulatory framework determines everything: what disclosures are required, how the product can be marketed, and what happens in the event of issuer insolvency.
Bybit has not clearly disclosed which legal regime applies to its tokenized equity product. That omission is itself a finding.
Regulatory arbitrage is not necessarily illegal. But it is risky for the end user. If the issuer is domiciled in a jurisdiction where token holder protections are minimal, then the entire structure rests on trust in a private contract. That is not how equities are supposed to work. Equities have clearing houses, depositories, securities investor protection funds, and a century of case law. Tokenized equities have none of those unless the issuer voluntarily subjects itself to such a regime.
Metadata does not mint value. The token is a claim, not the asset. The assertion that it is the same as the underlying security is marketing language, not legal language.
I want to see the SPV's organizational documents. I want to see the governing law clause. I want to see the dispute resolution mechanism. If a user in Singapore buys a tokenized Nvidia share issued by a Swiss SPV, can that user sue in Singapore? Or must they litigate in Switzerland? The answer determines the practical value of their claim.
3. Dividend Mechanics and the Corporate Action Problem
Corporate actions are the place where tokenized equities typically break.
Consider dividends. When Nvidia pays a dividend, the custodian receives cash. The issuer or SPV then needs to distribute that cash to token holders. This is an off-chain process. Every step involves a time delay, an operational cost, and the risk of error.
The critical question: does the dividend distribution occur automatically or manually? If manually, how often, and through what mechanism? If the issuer fails to distribute dividends, token holders have legal recourse only through the issuer's own jurisdiction. That is a long and expensive path, and for small token holders, it is practically unavailable.
Then there are stock splits. When Nvidia announced a 10-for-1 stock split in 2024, the mechanics of tokenized shares became particularly complicated. Tokens cannot automatically multiply without a smart contract upgrade. The mint-burn contract must be modified to reflect the new number of shares. If the contract includes a mechanism to issue additional tokens to holders, the protocol must verify who held the tokens at the time of the record date. On a public blockchain, this is straightforward in theory — the record date snapshot can be taken from the chain.
But corporate actions are not always that clean. Stock buybacks are a common occurrence with blue chips. If the company executes a buyback, the SPV's share count drops, but the economic value to token holders should remain unchanged. This requires accounting adjustments. Sloppy accounting can result in token holders being diluted or overpaid, with the difference appearing as profit for the issuer.
Stock lending is the most dangerous corporate action. Custodians often lend shares in their possession to short sellers under standard prime brokerage agreements. The SPV's custodian may be lending the very shares that back the tokens, generating yield on the collateral at the token holder's expense. You think you hold Nvidia exposure. In reality, the SPV has entered into a securities lending arrangement that exposes you to the counterparty risk of the borrower. If the borrower defaults, the cash collateral posted may not cover the value of the lent shares, and the token holder's claim is impaired.
This is exactly the kind of hidden liability that a forensic audit must identify. Based on my audit of RWA tokenization frameworks, I can tell you that securities lending clauses are almost never disclosed in the marketing materials. They sit in the fine print of the custodian agreement, buried between clauses about fees and liabilities.
4. Lending Products: The Collateral Paradox
Bybit's lending products create the most complex risk architecture in this entire offering.
Let me define what lending means here. A user can either deposit tokenized equities as collateral to borrow stablecoins, or lend their tokenized equities to other users in exchange for yield. Both directions have serious structural concerns.
For the collateral side, the loan-to-value ratio matters. Blue chips like Nvidia and Apple are less volatile than smaller caps, but "less volatile" is not "stable." Nvidia traded at a drawdown of roughly 40% from its 2025 high in the correction that followed GPU demand concerns. Tesla has traded with substantially higher volatility, routinely moving 10% in a single session. A 25% haircut on Tesla collateral is not sufficient to protect a lender in a fast-moving market.
I have modeled these scenarios. In the Compound protocol stress test I ran in 2020, I simulated a 40% drawdown on ETH collateral. The undercollateralization cascade was not linear. Liquidations triggered more liquidations as the price drop accelerated. Tokenized equities carry the same cascade risk, with a critical additional complication: settlement is not instant.
When a loan collateralized by a tokenized share defaults, the exchange must liquidate the token. Tokenized equity is not a stablecoin. It does not have constant deep liquidity. The token's liquidity is a function of how many users are willing to trade it at any given moment. In a market downturn, liquidity evaporates. A liquidation of a large tokenized position could push the token price to a discount relative to the actual share price, triggering additional margin calls across the entire lending book.
This is a cascading, systemic failure. It has happened in decentralized lending. It has happened in centralized lending. The new variable is that the collateral is not digital native but conceptually anchored to a traditional asset. That anchor should provide stability. In practice, it provides a false sense of security, because the token's liquidity and the oracle's behavior are both fragile.
The second direction of lending is no cleaner. If user A lends their tokenized Nvidia shares to user B, user A is taking credit risk on user B. Bybit acts as the intermediary, coordinating the loan and handling collateral. But the structure is not transparent from the outside. I want to know: are the loans overcollateralized or uncollateralized? If the borrower side is allowed to short tokenized equities, then loaned shares become liquidity for selling pressure on the token, which further destabilizes the product.
The most important question that Bybit has not answered: what happens when the price of the underlying stock moves against the collateral holder? Is liquidation triggered by the token price or by the reference price of the actual stock? If the reference price is used, the oracle becomes a single point of failure. If the token price is used, liquidation can be triggered by fluctuations in token liquidity rather than actual value of the asset. Neither answer is good.
5. The Oracle Problem
Based on my audit of tokenized bond infrastructure in 2025, I can state the oracle problem plainly: in any RWA system, the point where off-chain price data reaches the on-chain contract is the point of maximum risk.
For these Bybit products, the oracle must feed real-time Nvidia, Apple, and Tesla prices to the smart contracts that manage liquidation, lending, and valuation. The source could be a NASDAQ feed, a broker-provided price, or a third-party oracle network. Each has a different risk profile.
A stock market is not an always-on data source. US markets close at 4 PM Eastern. They are closed on weekends and holidays. During those windows, the on-chain price becomes a static snapshot. If Nvidia announces catastrophic guidance at 8 AM before the market opens, the token will trade on secondary markets at a price reflecting the new information, but the oracle may still report the previous close. Lending contracts that rely on the stale oracle will not trigger liquidations until the market opens and the oracle updates.
That gap is an exploitable window.
During the March 2020 market crash, the New York Stock Exchange imposed multiple circuit breaker halts. The token's oracle would have been reporting frozen prices while secondary trading continued. Anyone who understood the timing could have borrowed against overvalued collateral minutes before the oracle caught up, drained the lending pool, and vanished.
This is not a theoretical attack. It is a known vulnerability in any 24/7 trading market connected to time-bound traditional markets. I spent four days, during the 2017 whitepaper audit, cross-referencing the Paragon Coin roadmap against public domain technology releases. That taught me the value of timing analysis. Timing in markets, in oracles, and in settlement is the difference between a working product and an exploitable one.
Stress tests reveal what audits cannot. Static code inspection cannot capture the dynamic failure of an oracle during market hours, during circuit breakers, and during days of extreme volatility. I have always argued that stress testing, not code review, is the gold standard for evaluating collateral systems. I want to see Bybit's stress test output for a 30% single-day drop in Nvidia shares combined with a price oracle lag. I suspect they have not run that test.
The Terra Luna collapse in 2022 taught the industry a related lesson. The algorithmic stablecoin's failure was rooted in an incentive misalignment between the mint and burn mechanism. When the market turned, the mechanism accelerated the decline rather than cushioning it. A tokenized equity lending product with a stale oracle has the same structural flaw: it can accelerate a decline through forced liquidations that feed on themselves.
I compiled a 10,000-word timeline of the Terra collapse, interviewing three former developers and analyzing SEC filings. The most disturbing finding was how little the protocol's designers understood about the timing of their own mechanism. They knew the math. They did not know the market. Bybit's tokenized equity team faces the same knowledge gap: they may understand the product mechanics but not the market dynamics of 24/7 trading against a time-bound underlying asset.
6. The Regulatory Gray Zone
The regulatory environment determines the viability of this entire product.
Let me be direct about Bybit's status. Bybit is a global exchange serving non-US customers. It operates under a structure involving entities in Dubai, Singapore, and other jurisdictions. The company has faced regulatory action from several authorities, including the UK's FCA and India's financial intelligence unit. The status is complex, but it is not compliant in all major markets.
The tokenized equity product is launched for "eligible" users. Eligibility is determined internally by Bybit. The relevant question is whether the product is a security under any major jurisdiction's laws.
Under US law, a tokenized share of Nvidia is a security. The token is an investment contract because it represents an expectation of profit derived from the efforts of others. The SEC has jurisdiction. The fact that Bybit does not serve US users is legally meaningful but incomplete as protection — offshore entities that in fact target US persons have been the subject of SEC enforcement for years.
Under EU law, MiCA covers crypto assets. But tokenized equities might not fit neatly into the MiCA categories. They could be classified as financial instruments, which would require the token issuer to hold an investment firm license under MiFID. If the issuer is outside the EU, there are additional considerations.
Under Singapore law, MAS has made clear that tokenized securities must be issued under the Securities and Futures Act. The issuer needs a capital markets services license.
The issuer chosen by Bybit matters enormously. If the issuer is a Swiss-regulated entity, the product is backed by Swiss law. If the issuer is a US entity under SEC regulation, the product is backed by US law. If the issuer is a company in a low-regulation jurisdiction, token holder rights are significantly weaker.
Priors are cheaper than promises. I default to the assumption that an unstated structure is a weak structure. If Bybit's issuer is compliant and properly licensed, the announcement will be backed by clear disclosures. The absence of such disclosures is itself a finding.
The compliance checklist I have used since my 2025 RWA engagement is simple. Has the issuer registered with a securities regulator with actual enforcement power? Are token holders identified in the SPV's constitutive documents as beneficiaries? Is there a legal opinion on the token holder's direct claim against the SPV? Does the custodian recognize the SPV's interest in the shares? Has the auditor issued a report on the share count held in custody? Every question currently lacks a public answer.
7. Precedents and Failure Modes
This product has been launched before. The lessons are not being learned.
Binance's tokenized stock program launched in 2021, with tokens representing shares of Tesla, Coinbase, MicroStrategy, and Apple. The product operated for several months before being discontinued in July 2021 under pressure from regulators in Germany and other jurisdictions. The stated reason: Binance was offering securities without the appropriate licenses. The product had a clean interface, strong liquidity, and eager users. It still failed because the legal foundation was not built.
FTX had no tokenized equities product, but its collapse in November 2022 showed the dangers of centralized exchange structures where customer assets and exchange assets are commingled. The lesson: if the exchange holds the assets, the exchange's bankruptcy is a critical risk. Bybit's lending product means the exchange holds tokenized assets as part of the lending book. That is not the same as self-custody on the user's part.
Let me describe a scenario. A user deposits tokenized Nvidia shares into Bybit's lending contract. Bybit goes bankrupt — not a probable scenario, but not impossible. The user's claim is against Bybit's estate. The tokenized shares may be recoverable, but the process will take months or years. During that time, the user cannot trade, cannot borrow, and cannot access their collateral. This is the risk of centralized custody, regardless of the underlying asset's tokenized form.
tZERO is a permanently valuable case study. Overstock founded tZERO in 2015 to trade blockchain-enabled securities. The project raised hundreds of millions and never achieved meaningful liquidity. The problem was not the technology. The problem was that tokenized securities were illiquid: there was no secondary market. Users who wanted to exit found that the only liquidity was through the issuing platform, which had no incentive to provide a deep peg.
The same liquidity constraint could hit tokenized equities on Bybit. If Bybit is the primary venue for a tokenized Nvidia token, then the token's liquidity is a function of Bybit's order book. If Bybit faces a regulatory crackdown, or if the exchange suspends trading, the token's liquidity evaporates and the token trades at a discount to the underlying share. Users who believe they hold Nvidia will find that the token's price is not Nvidia's price.
Another practical failure mode is the redemption mechanism. In most tokenized structures, a user can either sell the token in the secondary market or redeem the token with the issuer for actual shares. Redemption is a manual process. It requires the user to complete KYC, submit the token, wait for the issuer to sell the SPV's shares or deliver them, and then settle. In a stable market, redemption is an administrative chore. In a market crash, redemption will be demanded by everyone at once. The issuer will not have enough operational capacity to handle mass redemptions.
This is the bank run problem applied to a tokenized equity product. And it is a bank run that the blockchain cannot prevent, because the redemption process is fundamentally off-chain.
The Paragon Coin whitepaper audit I conducted in 2017 taught me that the most dangerous documents are the ones that promise the most. Paragon promised voter verification, social impact, and a consensus mechanism that contradicted its own technical specifications. I spent four days cross-referencing their roadmap against public domain technology releases, and I found five critical contradictions. The lesson was simple: when the promotion outpaces the infrastructure, the infrastructure is usually fiction.
8. Liquidity and Market-Making Dynamics
The tokenized share's liquidity is a derivative of two factors: the underlying stock's liquidity and the token's adoption. The former is deep. The latter is shallow.
Nvidia trades hundreds of billions in shares every month across global venues. The token you buy on Bybit might see a fraction of that volume. The spread is likely wider. The depth is thinner. This matters for two reasons.
First, market efficiency. A tokenized share should track its reference price. In normal conditions, arbitrage between the token and the underlying stock keeps the token's price close to the stock's price. But arbitrage requires the ability to convert token to share or share to token in a timely manner. If the conversion is slow or expensive, the arbitrage may not happen, and the token can trade persistently at a premium or discount.
Second, liquidation events. In a liquidation cascade, the market needs buyers at liquidation prices. If the order book is thin, the impact cost of a large liquidation is severe. The liquidated position moves the token price far below the stock price, triggering more liquidations, which move the price further down.
I tested this pattern in my 2020 Compound analysis. Liquidation cascades are triggered not just by reference price movements but by the interaction between liquidations and order book depth. The collateral factor adjustments I flagged predicted the liquidity crunch in smaller forks. The same dynamic applies here.
The institutional users whom Bybit wants to attract will be particularly sensitive to this. An institution that allocates to a tokenized equity product wants both the asset exposure and the ability to exit. If the exit is uncertain, the product is not appropriate for institutional deployment.
I also want to examine the wash trading risk. In mid-2021, I analyzed the trading volume of a top-tier PFP project and demonstrated that 65% of reported volume was generated by five coordinated wallets. The raw volume numbers looked healthy. The unique active wallet counts told a different story. Tokenized equities could be subject to the same volume inflation, creating a false impression of liquidity. The presence of genuine liquidity cannot be confirmed without examining the order book distribution. Is Bybit market-making its own tokens? Or are independent market makers providing liquidity? The answer determines whether the volume and spread are economically meaningful.
9. Counterparty Concentration Analysis
Let me summarize the counterparty risks in one ledger.
The issuer: may fail operationally or economically. The custodian: may fail or commingle assets. The SPV: may not be bankruptcy remote. The broker: may fail to execute purchases correctly. Bybit: may face regulatory action or become involved in insolvency. The oracle: may lag or be manipulated. Each counterparty introduces a possible failure. Six counterparties means six failure modes.
For comparison, when you buy Nvidia shares through a US broker, you are subject to SIPC protection for up to $500,000 of securities. The broker's bankruptcy does not directly endanger your shares because brokerage accounts are held at DTCC and the shares are registered in your name. The tokenized equity product has no equivalent protection.
If the SPV is genuinely bankruptcy remote, token holders might survive the issuer's insolvency. But they will not survive a custodian's insolvency if the custodian commingled the SPV's assets with its own. They will not survive a legal determination that the tokens are not valid securities claims. The legal primacy of the token holder's claim is the entire question. No marketing material has ever provided a definitive answer, because none can guarantee the behavior of a court in a jurisdiction that has never ruled on tokenized securities.
Let me also address the question of audit trail. I have been told by RWA proponents that "the blockchain is an audit trail." That statement is technically true but practically misleading. The blockchain records the token movements. It does not record the share movements. It does not record the custodian's internal ledger. It does not record the SPV's accounting. The blockchain audit trail ends exactly where the off-chain world begins. And that is precisely where the risk lives.
Contrarian Angle: What the Bulls Got Right
I have spent thousands of words detailing the risks. A fair audit requires acknowledging where the bull case is sound.
Tokenized equities genuinely do improve access. The economic efficiency of holding a digital asset whose value tracks Nvidia or Apple, settled 24/7, without a traditional brokerage account, is real. For users in emerging markets — Southeast Asia, Latin America, Africa — a tokenized Nvidia share is a meaningful improvement over limited stock market access.
The institutional movement toward RWA tokenization is real. BlackRock, Franklin Templeton, and other asset managers are committing substantial resources. If the infrastructure matures, tokenized equities could become a standard component of the digital asset class. The direction of travel is correct.
Distribution through an exchange like Bybit is a pragmatic move. Exchanges have the order books and the user bases. The product's success depends on distribution and compliance; the former is clearly present. The latter is the open question.
The product's existence also pushes the regulated financial industry to consider more innovative structures. It normalizes the idea that securities can live on public blockchains. That normalization will eventually make settlement processes more efficient, reduce counterparty friction, and lower costs for retail investors.
I should also acknowledge that the mega-cap selection limits some risks. Nvidia, Apple, and Tesla are among the most liquid stocks in the world. The underlying assets are unlikely to become worthless. The risk is not in the stock. The risk is in the wrapper. A sound wrapper around a liquid asset can create genuine value.
But my role is not to validate the trend. My role is to identify the points of failure. The trend can be real and the product can be fragile simultaneously. The bull case rests on the assumption that the wrapper is sound. That assumption has not been demonstrated.
Takeaway: The Question Bybit Must Answer
The product is announced. The details are absent.
I want to see the custody agreement. I want to see the SPV's organizational documents. I want to see the mint-burn contract's permissioning. I want to see the stress test output for the lending book using a 200-day historical drawdown of each underlying stock. I want to see the oracle's behavior during scheduled market closures and circuit breakers. I want to see the legal opinion on token holder claims against the SPV.
Tracing the ledger back to the zero-day exploit would resolve the entire question. Until then, the product is a promise with a price tag.
Users who evaluate this product should apply the standard I have applied for eight years: can I verify the existence of the underlying asset, trace the custody chain, and exit my position without catastrophic loss in both normal and stressed market conditions?
If the answer is no, capital preservation is a prior. The market will not wait for you to understand the structure before moving against you. By the time the risk manifests in a liquidation cascade, the audit will be retrospective. And nobody was ever compensated for losing money because they trusted too fast.
I am not saying the product will fail. I am saying the evidence required to evaluate it has not been provided. In a bear market, where survival matters more than gains, that is not a detail. It is the entire story.