The code doesn't lie—but it also doesn't price a private company. Last week, Bybit added Unitree Robotics and Moonshot AI to its Pre-IPO perpetual futures lineup. The announcement was met with the usual bull market enthusiasm. I took a different approach. I pulled up the contract specifications, compared them to BitMEX's existing offerings, and ran the math on the pricing mechanism. The result is not a derivative innovation. It's a prediction market with a funding rate.
Context: What Is a Pre-IPO Perpetual?
A perpetual futures contract is a derivative with no expiry, relying on a funding rate to anchor its price to an underlying index. For crypto assets, that index is a transparent, liquid spot market. For Pre-IPO perpetuals, the underlying is a private company's valuation—a number that shifts only during funding rounds or secondary sales. BitMEX pioneered this in late 2024 with SpaceX, Stripe, and Anthropic. Bybit is now copying the playbook, but the choice of targets—Chinese robotics firm Unitree and AI startup Moonshot AI—adds a layer of opacity that even BitMEX hasn't tested.
Core: The Three Blind Spots in the Contract
I spent three hours reverse-engineering the pricing assumptions. Here's what the marketing glosses over.
1. Price Discovery: The Impossible Anchor
A perpetual contract's mark price must be updated frequently—every 5 seconds on most exchanges—to maintain the funding rate mechanism. For Unitree Robotics, the last public valuation was from a Series B in 2024, at $1.5 billion. That's a single data point, months old. Moonshot AI's last round was in early 2025, with a reported $3 billion valuation. These are not continuous prices; they are discrete events. Bybit must either use a stale price or extrapolate jumps based on news. Neither is a reliable anchor. In my 2020 Uniswap V2 deconstruction, I saw that AMMs require continuous arbitrage to maintain price accuracy. Here, there is no arbitrage because there is no liquid spot market. The contract becomes a standalone speculation tool, not a derivative.
2. Funding Rate: A Mechanism Without a Market
Funding rates require traders to bet on the convergence of futures and spot prices. Without a real-time spot price, the funding rate cannot converge to a rational level. It will oscillate based on sentiment, not arbitrage. Based on my analysis of the Gnosis Safe code in 2018, I learned that trustless systems require verifiable inputs. Bybit's funding rate is based on an internal index that lacks public verification. This is a centralization risk that no audit can fix.
3. Settlement: The IPO Dependency
The contract likely settles when the company IPOs—or converts to a stock-related contract. But what if the IPO is delayed by a year? Or canceled? The contract becomes a zombie position. Bybit's terms (not publicly disclosed in full) likely include a forced settlement mechanism, but the trigger conditions are opaque. This is a liquidation risk that traders cannot hedge.
Contrarian: The Real Threat Is Not the Contract—It's the Valuation Blindness
Most critics focus on the lack of regulation or the small market size. I see a different problem: the pricing model is a black box. Bybit is likely using a third-party data provider (e.g., PitchBook or a private secondary exchange) to generate a valuation index. But these indices are themselves estimates based on limited data. For Chinese companies like Unitree and Moonshot AI, the secondary market is almost nonexistent. The index will be heavily influenced by media reports and rumor. This is not a decentralized oracle; it's a centralized opinion feed. In the 2021 Axie Infinity forensics, I found that smart contract bugs can be patched, but economic design flaws are permanent. This pricing model is a design flaw.
Takeaway: The Vulnerability Forecast
I don't believe these contracts will attract meaningful liquidity. The funding rate will remain erratic, and the mark price will diverge from any rational valuation. When the next bear market arrives, these Pre-IPO perpetuals will become toxic assets—illiquid, mispriced, and impossible to unwind. The code may be clean, but the math is broken. Zero knowledge isn't magic; it's math you can verify. Bybit's Pre-IPO perpetuals don't pass that test.
