TechFlow Logo
Login/ Sign up
ETH Gas
Gwei
Fear
gas
Deconstructing the Next Crypto Cycle: VC Will Disappear, Prediction Markets Overvalued, Which Perp DEX Can Challenge Hyperliquid?

Deconstructing the Next Crypto Cycle: VC Will Disappear, Prediction Markets Overvalued, Which Perp DEX Can Challenge Hyperliquid?

2026.08.13
Share
From Primary VC to Cross-Asset Trader, After Putting Down the Blockchain "Hammer".

Authors: Victor, Mr. Z,168X

August 2026, Crypto is at the bottom of the industry cycle: the market is in a severe bear phase, exchanges like BitMEX and BitMart announce they are halting trading operations, star products from the last cycle like Zapper and Fantasy Top are gradually shutting down, and DEXs and DeFi protocols continue to face hacks. More critically, talent and capital are flowing massively into AI, and primary market VCs are beginning to liquidate their positions. At this moment, 168X invites Lao Bai (@Wuhuoqiu) to discuss why Crypto has clearly won, yet everyone feels like they are losing? Where we stand now, and what opportunities remain for the next cycle.

Lao Bai has ten years of experience in network engineering and entered the crypto space in 2017 to buy OriginTrail. He later worked in research and investment at Amber, ABCDE, and OKX Ventures. His research scope now spans AI, semiconductors, and the tokenization of traditional finance. He has narrowed his focus within crypto to tracks that still have PMF: stablecoins, perpetual futures (Perps), RWA, and prediction markets. In this nearly two-hour conversation, Lao Bai makes a series of sharp judgments: issuing tokens is essentially debt, not fundraising; the entity known as "Crypto VC" will disappear; the ceiling for prediction markets is far lower than that for Perps. He believes the era of treating blockchain like a hammer where everything looks like a nail is over. The two strongest inventions in this track are not a new public chain, but stablecoins and perpetual contracts. And Robinhood represents the ultimate form of all future exchanges.

I. Lao Bai's Investment Evolution: From Engineer, Primary Market VC to Cross-Asset Researcher

Victor: First, please introduce yourself, Lao Bai. Share your experience with everyone, and tell us which tracks you are currently focusing on?

Lao Bai: Thank you, Victor, and thank you, 168X. My previous career path involved starting as a researcher at Amber, then becoming an investment research partner at ABCDE, followed by a few months in investment at OKX Ventures.

Personally, my focus has actually become quite narrow. Within Crypto, it has converged to a few tracks: Perps, prediction markets, and RWA. These are among the few areas where I believe there is still PMF (Product-Market Fit). The rest of my energy and time has expanded into US stocks, especially AI-related ones, and some other investment opportunities. Recently, I've been focusing heavily on options because they greatly expand how I can express views on an asset's price or value. So I've been researching this area extensively.

II. Blockchain as a Hammer: Why Most Star Tracks Are "Just for the Sake of Going On-Chain"

Victor: From 2017 to now, we've gone through several cycles. You previously wrote an analysis of Web3 gaming, mentioning "we once treated blockchain like a hammer, making everything look like a nail." Many things in the world fundamentally don't need to be financialized. Based on your past VC and investor experience, how do you view the development and evolution of these projects across the recent cycles?

Lao Bai: I'd like to elaborate on this; it's an excellent question. Because when I was doing primary investments myself, I had this exact feeling: whether VC or founder, many of us make the same mistake—treating blockchain like a hammer, so everything looks like a nail.

If you start from first principles of blockchain, setting aside terms like decentralization, privacy, and censorship resistance (which 99% of users actually don't care about), the core value of blockchain is exactly what Li Xiaolai or Lao Mao wrote about in a 2017 article: blockchain is the world's first network to achieve peer-to-peer value transfer. The internet allows us to transmit information, while Bitcoin, or rather blockchain, allows us to transmit value, not just information. How does it achieve this? Simply put, it prevents infinite copying (copy-paste) of information through technology like preventing double-spending. You could even imagine blockchain as an evolutionary version of BitTorrent, but instead of a peer-to-peer download network, it becomes a peer-to-peer value transmission network. Then, a very important feature is: assets on it can be programmable.

If you dig into Satoshi's earliest posts around 2009 to 2012, you'll find his original vision was actually to carry more assets on top of Bitcoin after its success, including invoices, bonded contracts, etc. So Bitcoin originally intended to do what Ethereum later did, but Satoshi stepped back, and Vitalik Buterin emerged out of nowhere to realize Bitcoin's initial idea with Ethereum.

This is also why I was particularly bullish on OriginTrail back in 2017. At that time, Changxia's concept was about bringing real-world assets on-chain. I thought that's what blockchain should do, so the first altcoin I bought was OriginTrail. But looking back now, OriginTrail belonged too early—it became a pioneer martyr rather than a trailblazer. By late 2020 and 2021 during DeFi Summer, with asset prices skyrocketing, I got carried away and forgot the original intention of "bringing assets on-chain."

But by 2024, I started feeling something was off. I discussed this in an internal meeting at ABCDE: After counting them up, the dozens of tokens I personally consider valuable are all focused on enhancing the programmability of the entire blockchain network. From Uniswap evolving from V1 to V2, V3, Aave doing the same with V1, V2, V3, to later projects like Morpho and Pendle, we have continuously enhanced programmability, as well as the speed and capacity of the entire network—essentially the infrastructure layer. But where are our assets? I found almost none. We tried NFTs, GameFi in-game items, inscriptions, Ordinals, and even current Memes, but ultimately found they lack long-term asset attributes and value.

Regarding in-game items, Vitalik promoted this concept earliest. Back in 2017, he mentioned that he believed finance and gaming were the first two scenarios that could be successfully implemented. As a result, we thought about putting game assets on-chain as a type of RWA. As expected, it has already failed. So half a year ago, I wrote an article titled "Why World of Warcraft Succeeded While GameFi Failed." My point is: Anything that can be scaled can definitely be industrialized, and it contains an impossible triangle. Even if you allow assets in World of Warcraft to circulate freely like on a blockchain, the game would be destroyed by farming studios, let alone our chain games with zero playability.

Speaking of star tracks across different cycles, I think there are a couple of categories. One category, like OriginTrail, arrived too early and became pioneers martyrs, including Augur, Ethereum's earliest prediction market, which also suffered from being too early with wrong timing. Another category is where execution didn't catch up, like the earliest DEXs, Bancor and Kyber, which were later directly front-run by Uniswap.

Victor: Bancor seems to have shut down recently too.

Lao Bai: Yes, this token can basically be considered a failure. Everyone already assumes it's a failed project, so whether it shuts down or not doesn't matter much. So now when I evaluate projects, I basically try to distinguish: Is it blockchain for blockchain's sake, or is it using blockchain to serve business purposes and reduce costs while increasing efficiency? The simplest way is to remove the token and see if your product actually works. You'll find that Uniswap works, Aave works; but GameFi projects like Axie Infinity and STEPN completely don't work. So for projects that fail without the token, we can now collectively consider this not a correct development direction. You ultimately need to figure out what task blockchain solves that traditional finance cannot, and then see how it reduces costs and increases efficiency in the real world.

III. Why the Token Issuance Model Fails: From "Too Much Meat, Too Few Monks" to "Too Many Monks, Too Little Meat"

Victor: So when you evaluate projects now, you mainly look for PMF and whether they can generate cash flow. In the past, token issuance was often a model that allowed VCs, retail investors, and various industry roles to participate. Why did it work before but isn't working now?

Lao Bai: Before, token issuance worked because the industry was at a very early stage, allowing everyone to assign a high "price-to-dreams ratio" valuation. Even if you have no users and no revenue right now, as long as KOLs and VCs believe your business model can work over the next five to ten years, we're willing to give you a high valuation. Retail investors' pursuit of these tokens follows a completely different logic than stock investing. It's entirely based on following the judgments of KOLs and various VC institutions. For example, if you raise money from a16z or Paradigm, retail investors accept the institutional viewpoint. After the project issues tokens, what retail investors actually buy is similar to a "price-to-dreams option" for the next five to ten years.

The second reason is that there weren't enough projects on the market at the time. Hundreds of billions, even hundreds of billions of dollars, flowed in, but we only had dozens, or a hundred or two projects. This is a classic case of too much meat and too few monks. But this state couldn't last long. As more and more project teams entered the market and more VCs were established, everyone started making a PPT to seek VC funding and endorsement, then used high FDV, low circulating supply, and locked distributions to control the market, making retail investors provide the final liquidity. When this crossed a critical threshold and turned into too many monks and too little meat, the entire cycle was broken. Liquidity couldn't sustain it, leading to the current situation.

Victor: Many communities still discuss whether projects with strong cash flows, like Polymarket, even need to issue tokens. People say OpenSea was surpassed by Blur and gradually declined precisely because it didn't issue tokens at the optimal time. Do you think this logic holds? Could Polymarket tap a larger market and surpass Kalshi by issuing a token?

Lao Bai: From a retail perspective, you could say Blur succeeded at the time and OpenSea failed because it didn't issue tokens at the right moment. If it had, the valuation would have been at least tens of billions. But from Polymarket or OpenSea's own perspective, with such strong cash revenue, issuing tokens is unnecessary because fundamentally, the token you issue is actually your debt. For the project team itself, issuing a token doesn't necessarily maximize their financial or brand commercial value. Moreover, Blur isn't really a thriving project now either; it only briefly surpassed OpenSea during that period.

Polymarket certainly could surpass Kalshi in traffic and other aspects by issuing a token. But taking recent months as an example, opinion搞ed a token airdrop wave from late last year to early this year, with trading volume nearly matching Polymarket. Yet after issuing the token, both trading volume and token activity have almost hit zero. So from my perspective, regarding Polymarket, they don't actually have a rigid need to issue tokens. Whether they do or don't is fine.

IV. Crypto Clearly Won, Why Do Veteran Retail Investors Feel Like They Lost?

Victor: There's a current situation: many VCs and retail investors feel Crypto has already won. Bitcoin has entered mainstream society, launched ETFs, and Wall Street increasingly acknowledges it. But why do veteran retail investors within the industry seem to feel they lost, not making money like in past cycles? Is it because Crypto is being gradually absorbed by traditional finance, leaving less participation opportunity for retail investors?

Lao Bai: That's indeed true, and I believe both points hold simultaneously: First, Crypto has matured. Second, Crypto's original ambition to transform the world has indeed shrunk. In other words, we started wanting to natively create a new Crypto world, but slowly degenerated into reformers and vassals of the existing financial system. It feels a bit like the Lianshan heroes being granted amnesty at the end of Water Margin.

If you take a time machine back seven years and told VCs, founders, or retail investors from the 2018-2019 era: Stablecoins have now become a very crucial global USD infrastructure with a massive scale of hundreds of billions; Bitcoin has ETFs, with pension funds and institutional capital buying in; Crypto companies can participate in IPOs; RWA is truly entering the traditional financial system, with companies like Stripe, Visa, and BlackRock deploying Crypto infrastructure, building public chains, and creating PayFi. People from that era would definitely feel: We've already won!

But looking from an asset perspective, we indeed haven't won. Another DeFi Summer, or altcoins pumping 50 to 100 times in a year—that angle simply isn't there. And over the past decade, the endeavor of natively issuing assets on-chain, from NFTs to inscriptions to various Memes, I believe we lost that battle too. Ultimately, we only built a rail for on-chain finance, and now we're starting to bring genuinely native real-world assets onto the chain.

However, we have two incredibly brilliant inventions: one is stablecoins, and the other is perpetual contracts. These belong to what I consider a complete victory, a truly Crypto-native PMF.

Mr. Z: I resonate deeply with your remarks. A few weeks ago, I saw Haseeb, managing partner at Dragonfly, say that Crypto is gradually maturing, and he believes the concept of "Crypto VC" may not exist in the next decade, just like how we invested in IoT and social platforms like Facebook in the mid-2000s, and today no VC claims they are investing in those specific things. So do you think Crypto VCs will cease to exist in five or ten years?

Lao Bai: I basically agree with Haseeb's view. In the future, whether as a VC or even a Trader, you shouldn't treat Crypto as a separate industry or sector to worry about. Just like you wouldn't call a VC an "Internet Fund VC" today—the Internet serves as a technology to support the commercial system. Crypto will be the same in the future; it will serve as a technological foundation, integrated into your entire business model. When you need to use stablecoins or put certain things on-chain, you might allocate only 10% to 20% to commercial projects or companies within Crypto. It's hard to still call yourself a "Crypto VC" or label something a "Crypto project." I believe Crypto should gradually integrate into the real world like the Internet, and people will no longer need to worry about whether it's a Crypto asset, Crypto project, or Crypto VC. These labels will disappear.

V. Perp DEX Competitive Landscape: Four Little Dragons and HIP-3, Who Can Challenge Hyperliquid?

Mr. Z: This really aligns with the current operations of Perp DEXs like Hyperliquid and Variational. They use blockchain as a base to move assets on-chain, for example, using protocols like HIP-3, allowing deployers such as TradeXYZ and Paragon to bring assets on-chain, enabling users to trade stocks from various countries 24/7. So what do you think the competitive landscape for Perp DEXs will look like in the next two to three years?

Lao Bai: First, I am personally very bullish on Variational. Outside of Hyperliquid as the Tier 0, my preferred second tier ranks first as Aster, second as Lighter, and third as Variational. For formats like Variational combining RFQ (Request For Quote) with OTC hedging, the logic makes perfect sense. Combined with the excellent team, founder, backing capital, and institutional resources, it's highly promising.

I believe the market only really needs four or five Perp platforms, just like exchanges only need four or five. Previously, it was Binance and OKX, plus a few others users liked like Bitget, Bybit, MEXC, and Gate, totaling just a handful. In the end, Perps will likely consolidate into four or five leaders forming a dominant effect: Tier 0 is Hyperliquid, and Tier 1 includes Lighter, edgeX, Variational, etc.

A few days ago, I chatted with a friend who raised a very interesting point. As Crypto Perps and CEXs increasingly resemble leveraged contract versions of traditional finance and stock brokers, many small-to-mid-cap stocks with average liquidity might adopt the Binance Alpha playbook in the future. The strategy involves controlling 60% to 70% of the float at the spot level, continuously buying to pump the price, while simultaneously crafting a narrative—like a traditionally struggling company suddenly announcing a pivot to AI or DAT (Digital Asset Treasury). At this point, a very high Open Interest (OI) might accumulate on Hyperliquid or Binance. Then, the company's shareholders or stakeholders can exit through the Perp market. You can't find any regulatory loopholes at the spot level because the company is only repurchasing its own stock, buying and never selling, so regulators have nothing to say; but the actual selling happens inside the Perp market. This is a long-term regulatory arbitrage. Coupled with our Perp DEXs, it creates a new gameplay mechanic for other liquidity-average small-to-mid-cap stocks globally.

Mr. Z: I observe that successors beyond Hyperliquid actually need alternative strategies. Hyperliquid already lists many US stocks; everyone should have standard US stocks like optical interconnects and storage by now. Could successors like Variational explore Asian stock markets, such as listing TSMC, MediaTek, UMC from Taiwan, or Alibaba, Tencent, Meituan, JD.com from Korea, Japan, and Hong Kong? Because going head-to-head directly with Hyperliquid is a losing battle. How do you view this aspect?

Lao Bai: Perps like Lighter and edgeX, which share the same architecture as Hyperliquid based on CLOB and market makers, indeed should pursue differentiated competition. The Asian assets you mentioned, TSMC, MediaTek, Korean stocks, these directions are all correct. I remember Mable's Trasia is exactly heading in the Asian asset direction. Another friend of mine also plans to leap from VC to build their own Perp, targeting the on-chain forex direction. We've shouted about this for many years but never made it happen. If stablecoins on-chain reach a scale of trillions, the Total Addressable Market (TAM) for on-chain forex is actually enormous.

But projects like Variational, including Ondo Perp, are different. If you look at their slippage and liquidity, they can achieve better, tighter spreads than even Hyperliquid or Binance. Because Hyperliquid essentially relies on a single market maker to provide liquidity, while Ondo operates similarly to an RFQ quote, allowing it to hedge using its own TradFi capabilities, locking and hedging extremely conveniently within the venue. Therefore, on many RWAs, especially large caps like NVIDIA and Micron, the spread and slippage of Ondo Perp are superior to Hyperliquid and Binance, which I verified myself. Theoretically, Variational's mechanism can also replicate Ondo's setup. Moreover, its swap launch last month introduced traditional TradFi hedging mechanisms for commodities like crude oil and gold. So I believe these two have different mechanisms and are worth considering to go head-to-head with Hyperliquid.

Mr. Z: Looking into Hyperliquid specifically, within the HIP-3 segment, TradeXYZ is still the leader, but there are rising stars like Paragon with quite clever strategies. They emphasize speed and efficiency, quickly auctioning and snapping up tickers to launch assets wherever market hotspots appear. They don't build their own frontend UI, believing Hyperliquid's interface is already good enough. Have you been paying attention to this?

Lao Bai: I'm paying attention, but maybe not to a huge extent. On the Hyperliquid frontend now, I can only see deployers like TradeXYZ and Paragon. Some assets on Paragon differ slightly from TradeXYZ, but currently, its volume is still quite small, mostly in the hundreds of thousands range, while TradeXYZ consistently handles ticket sizes in the hundreds of millions. The gap is massive.

Originally, that HIP-3 ticker merely marked a deployment opportunity. The first three slots are free, and subsequent ones are bid for via auction. What asset you choose to run with that slot is entirely up to you. But I think this will largely result in a leader effect, and it's not solely an issue of user mindshare. Because market making isn't handled by the Hyperliquid official team; it's done by TradeXYZ and Paragon themselves. For the same underlying asset, if TradeXYZ's market makers execute liquidity and spreads exceptionally well, everyone will naturally gravitate toward TradeXYZ. This creates a self-reinforcing growth flywheel. Breaking this loop is truly what successors need to figure out. Moreover, unlike Uniswap where you can utilize new AMM mechanisms—like Curve did with stablecoins by using different curves to achieve different effects—HIP-3 purely comes down to competing on market makers' liquidity and depth. It's simply a比拼 of who has more real money, making it difficult to innovate cleverly in this space.

VI. Security is a Function of Time: Lessons from Ostium, Cream, to Perp

Victor: Recently, there have been some incidents in the Perp space. TradeXYZ faced a SK Hynix wick event recently; a similar thing happened on Binance but with minimal impact. TradeXYZ ultimately chose to compensate all users for their losses, which was very generous. Other projects have also been hacked or experienced bugs lately, such as the RWA project Ostium being hacked for approximately $18 million recently, and Paradex getting hacked not long ago. How will these security issues influence the hesitation surrounding DEX security, and whether Perps will gain broader acceptance from traditional finance and investors?

Lao Bai: I think this depends on two points. First, it requires time to prove itself, because 99% of users won't possibly audit your underlying code; they rely purely on reputation and time. Why is AAVE so strong with such high consensus? Because it's a lending protocol that has never suffered a security breach. Even the previous incident was actually caused by underlying assets compromising the protocol, not an AAVE-specific flaw.

You'll find that the vast majority of security attacks on lending platforms occur precisely because of the lending mechanism itself: using a junk asset as collateral to drain massive amounts of stablecoins by manipulating oracle feeds, pumping the price, or controlling voting power. Platforms like Mango on Solana at that time, and many other junk lending protocols, suffered attacks, sometimes multiple times. Take Cream, the lending platform by Taiwanese influencer Machi. I remember it was attacked three or four times. I never understood how anyone could still bravely deposit money into it after being hacked.

So regarding Perps, I think there are two main points. First, do not expand your Perp business to the point where you can mortgage certain assets and integrate lending into it. I remember Paradex, or rather Drift on Solana, wanted to be comprehensive, doing spot trading, collateralized lending, and Perps all at once, ultimately getting breached through the lending side. If you build a Perp, just honestly build a Perp. Don't mess around with collateralizing assets to extract stablecoins. Once you do that, a breach is inevitable eventually. Second, there's no other way but to rely on time to prove yourself. If four or five years pass and Hyperliquid has never experienced any security incident while competitors have, users will naturally trust it and be willing to park their funds there, just like how Binance completely dominates user mindshare.

VII. Will Hyperliquid Launch Spot Trading? The Ideal vs. Reality of Tokenized Stocks

Victor: Regarding spot trading, Hyperliquid hasn't launched it yet. Do you think there's a possibility they will enter the spot market? To keep up with the strategy you mentioned about Binance Alpha potentially launching assets for stocks, would they need to launch spot trading?

Lao Bai: I think they won't do spot trading for native Crypto. Spot trading's impact is simply too small right now. Look at Binance, where 80% to 90% of its revenue comes entirely from Perps. And within those Perps, starting last month, over 60% of Binance and HTX's volume already comes from US stocks, i.e., RWAs, rather than native Crypto-based pairs.

So if you want to list spot assets, you either list spot stocks, similar to Binance's bStock. I'm uncertain if Hyperliquid will eventually launch something like "hyperstock." The possibility exists, though slim. But if they do, the sole purpose would primarily be to service their Perp market. You could perform cash-and-carry arbitrage or conduct on-exchange hedging. Take Ondo: its Perp liquidity is so robust precisely because it holds its own Ondo Stocks internally. When a market maker quotes you and you open a $1 million long position on Micron, they can simultaneously offer you a short position as a counterparty, directly holding $1 million in Micron spot inventory with zero latency. So I wouldn't be too surprised if Hyperliquid takes this action, but they absolutely will not launch spot trading for native Crypto assets like BTC, ETH, or SOL.

Victor: Between tokenized stocks and Perps, if you had to choose only one, which is the better model? Or are they complementary? Platforms like StableStock position themselves as stablecoin brokerages, letting users buy stocks directly with stablecoins but without Perps. Meanwhile, Binance offers a full suite with Perps, US stocks, and tokenized stocks.

Lao Bai: The ideal state for tokenized stocks should be "real stocks combined with clear legal ownership plus free on-chain circulation." Interestingly, Nasdaq itself might actually be moving toward this. I recall the SEC approved Nasdaq's initiative to launch on-chain stocks a couple of days ago. So regarding stock spot trading, the pricing competitor for StableStock, including my partnered platform BIT, might actually be Nasdaq itself. It's just that Nasdaq likely won't allow you to freely buy and sell using stablecoins in the short term.

But from a practical standpoint, the vast majority of users don't actually require clear legal ownership backed by real stocks. They simply need a price exposure to bet on, as long as that exposure anchors to the real stock price. Therefore, the most mainstream format remains Perps. This is exactly why I stated that over 60% of revenue now comes from Perps, and within Perps, another 60% is driven by RWAs. Ideally, everyone would be holders engaged in long-term investing like in traditional US stocks; but realistically, everyone is a gambler seeking exposure, adding 5x or 10x leverage on that exposure, crazily opening longs and shorts. Ideals are ideals, reality is reality.

VIII. Stablecoins: Crypto's Greatest Invention and the Global Extension of Dollar Hegemony

Victor: Next, let's talk about stablecoins. Recently, projects like Open USD have gathered major traditional financial institutions like Visa, Stripe, and Mastercard, reminiscent of Facebook's Libra back in the day. How do you view the evolution of the stablecoin track in the next cycle? If new projects are to emerge, what kind of differentiated advantages would they need?

Lao Bai: I am infinitely bullish on stablecoins, super bullish. I believe this is the most important invention outside of Bitcoin, period. Crypto's greatest contribution to the world is actually stablecoins, even more so than Bitcoin. Bitcoin is currently mainly a game for a small number of whales, institutions, and giants, but stablecoins are the technology that truly changes the world via blockchain.

Last year, we were discussing using stablecoins to monetize national debt. The US national debt was $36 trillion last year, and it's already $39 trillion now. Everyone is thinking that stablecoins should urgently reach one or two trillion dollars before 2030, which could at least help offset two trillion dollars of US debt, alleviating the burden on the US government. In another ten or twenty years, it might scale to five or ten trillion. You can see that on the B-end, heavyweights like Stripe and Visa are jumping in across scenarios ranging from payments to yield generation; on the C-end, it goes without saying that stablecoins enable the buying and selling of virtually any asset globally. Therefore, ultimately I believe stablecoins and tokenized US stocks act synergistically: dollar hegemony combined with dollar asset hegemony penetrates every country and corner of the world through stablecoins.

As for wanting to enter as a new player, I think the barrier is extremely high. New entrants can only be super players like Stripe, BlackRock, and Visa, which already control traditional payment gateways, fiat financial entry points, or intermediary settlement channels, thereby earning the right to carve out a share of the stablecoin business. Regular entrepreneurs are advised not to touch this space; this is a highly institution-dependent business.

IX. Agent Payments and Machine Economy: Cloudflare's Traffic Inflection Point and Micropayments

Victor: The narrative around stablecoin and AI Agent payments was very popular last year, like x402, as well as the recently launched agent wallets from Cloudflare and Tether itself earlier this year. All aim to emphasize Agent payments for stablecoins and the machine economy. Is this a viable narrative or more of a dream?

Lao Bai: This narrative is highly viable, but whether it has anything to do with Crypto is debatable. Let's look at it from two angles.

First, will the agent economy actually happen? I believe it will. Last month, during an interview, Cloudflare's CEO mentioned an event: On a specific day last month, bot traffic on the world's internet surpassed human traffic for the first time. Occurring in July, it's a massive milestone. The internet has always hosted numerous bots, but human traffic historically dominated. Now, acting as the world's largest gateway/CDN, Cloudflare's data shows bot traffic finally exceeded humans. The reason is Large Language Models: If a human wants to buy a DSLR camera, they might check five webpages to compare prices, generating five HTTP requests based on those five visits; but if you ask ChatGPT or Doubao "Recommend a $700 DSLR camera with these specific features," to give you the optimal answer, it might generate 5,000 HTTP requests, browsing all relevant e-commerce and review sites online. Thus, the machine era demands exponentially explosive network traffic. Cloudflare's CEO predicts agent traffic will expand a thousandfold over the next five years.

Second, he raised a question: Previously, the internet's economy relied on Google Ads; we completed the entire commercial loop based on search and recommendations through ad clicks. But once an agent recommends something to you, that model is broken. The only visible solution right now is implementing micropayments when an agent scrapes your website's data. For instance, if you run an e-commerce or review site, and an agent utilizes your data across 5,000 scrapes, it pays you a couple of cents. How is this paid? Likely through mechanisms like x402, or slowly rolled-out stablecoin-based solutions by Cloudflare, Visa, and Stripe. It definitely won't be traditional Visa, nor necessarily pure Crypto, but it will absolutely involve stablecoins running on some chain. Whether that chain is Solana, Ethereum, Tempo, or Arc, we don't know, but I believe it will inevitably unfold in such a scenario.

X. The Endgame of Public Chains: Universal Chain Era Ends, Split Between Institutional Chains, C-End Users, and Gray Markets

Victor: How will stablecoins and agent payments impact the competitive landscape of public chains? Currently, Ethereum is still the largest, alongside Solana, Circle's Arc, Stripe's Tempo, and Plasma from last year, etc.

Lao Bai: In the next cycle, I feel the scenario will largely be "Render unto Caesar what is Caesar's, and unto God what is God's," with everyone advancing side by side. If engaging in gray-market activities, people will likely still choose TRON and USDT; for retail investors, they'll mostly pick USDC on Ethereum and Solana to play DeFi and Memes; if you are an institution, B2B enterprise, or handling corporate payments/forex conversion, you'll predominantly use institutional chains like Tempo, Arbitrum, and Canton (a privacy-focused chain that has also developed excellently). It's highly probable that three or four institutional chains will serve the B-end, Ethereum and Solana will serve the C-end, and TRON will cater to gray markets. I believe this will be the dominant landscape.

Victor: In past cycles, everyone treated public chain tokens as the most mainstream Beta, like Ethereum and Solana. Will public chains remain the main axis for VC or retail investment moving forward?

Lao Bai: Definitely not. Hyperliquid has basically set a benchmark for all public chains: public chains must serve applications. I exist as a chain itself, but only to serve an incredibly powerful application, rather than aiming to build a universal public chain. The era of universal public chains likely reached its peak after Ethereum and Solana, but we refused to accept it, leading to launches like Aptos, Sui, Monad, and MegaETH. Look at Shuyao from MegaETH recently tweeting that their MegaMafia incubation program will cease funding new projects on-chain, deciding to dive in themselves to build applications. This signals that the universal chain era can be declared over; the future will undoubtedly be an application chain era.

This closely resembles AI: First, you build infrastructure. We had to lead by a generation, get the infrastructure solid first, resulting in sky-high infrastructure valuations. Now, people finally start prioritizing applications, PMF, users, and cash flow, and the crypto space is gradually mimicking the US stock market. Ironically, US semiconductor and storage sectors are somewhat mirroring our previous crypto cycle, starting to "crypto-fy," which is why many investors got wiped out leveraging 2x long on SK Hynix.

XI. Large Models Going "Chain-Like", Killer Apps, and Insights from Palantir

Victor: Currently, large models are somewhat like public chains were back then. The barrier to building large models keeps dropping, and the main battlefield shifts to specialized models tailored for specific applications. Do you think large models will follow a development path similar to public chains?

Lao Bai: I believe they already are. Look at China's "Six Little Dragons"—Kimi, DeepSeek, MiniMax—they already evoke that public chain vibe, where every large model has its own distinct characteristics. MiniMax's latest model, for example, rolls out an uncensored version immediately, delighting many video creators. So we've already carved out unique traits in technology, cost-effectiveness, and ethical review constraints, resembling the great public chain debates of yesteryear.

Ultimately, breaking through comes down to two factors. First, who can better integrate into the B-end. Both OpenAI and Anthropic have formed frontline deployment engineer teams themselves, emulating Palantir's approach, aiming to deploy AI directly within enterprises. Because as you see, despite our hype, over 90% of companies only equip employees with access to Anthropic, OpenAI, or Doubao, while the entire corporate workflow and architecture remain stuck in traditional legacy systems. AI's uplift for overall enterprises might be less than 20% to 30%, but for individuals, it could exceed 500%. Hence, OpenAI and Anthropic are eager, deciding to step in personally and establish dedicated companies to help enterprises seamlessly weave AI into their workflows. Second, for the C-end, whoever first builds a killer app beyond ChatGPT, Codex, and Claude will win.

Victor: A few days ago, Palantir's CEO also mentioned in an interview that the narratives Anthropic shares with the public are actually things Palantir has been doing all along, which is quite funny.

Lao Bai: Exactly, which explains why Palantir's stock has surged recently. People realize these highly capable SaaS platforms aren't easily replaceable by large models, and integrating LLMs into a corporate workflow isn't as straightforward as assumed.

XII. The Truth About Prediction Markets: Completing the Gartner Curve in Two Years, Ceiling Far Below Perps

Victor: Finally, let's discuss prediction markets, which attract significant attention. You've also invested in several prediction market projects, like 42space backed by YZi Labs. How do you view the current landscape of Polymarket, Kalshi, and prediction markets as a whole?

Lao Bai: Right, I just posted a thread a couple of days ago outlining the four stages of prediction markets: undervalued before elections, validated by Kalshi and Polymarket post-election, overvalued before the World Cup, and now beginning a return to rationality. You can observe that Polymarket's volume has been steadily declining since the World Cup kickoff on July 18th. So now, if Polymarket were to achieve a valuation around $20 billion, or actually issue a token reaching a $20 to $30 billion level, I would short it without hesitation. Because based on my observations, prediction markets operate on a completely different magnitude compared to Perps.

Back then, we projected many irrational imaginations onto prediction markets, somewhat like "looking for nails with a hammer." While at ABCDE, I discussed over ten prediction market concepts, and everyone brainstormed scenarios like friends betting amongst themselves using a private server room on a prediction market to place bets separately. But in practice, the frequency and scenarios of such friendly bets are incredibly low. It's fundamentally not a PMF; it's a fabricated PMF forced upon the concept just to justify building prediction markets.

Simply put, an individual doesn't constantly need to express opinions on countless events while simultaneously wagering on them. For the vast majority of people, it's likely restricted to politics and sports. For sports, traditional bookmaking already satisfies demand exceedingly well. In Europe, America, and Australia, people love attending stadiums to watch games or heading to sports bars, where crowds drink beer, watch horse racing or matches, and casually buy a traditional sports lottery ticket for a few dozen Australian dollars. Winning brings joy, losing costs only a meal. This is precisely why Kalshi performs decently in sports, but essentially, the underlying demand for these has already been largely satisfied by traditional markets and bookmakers.

Recently, I spoke with a founder building a prediction market tool—a platform akin to "GMGN for prediction markets" featuring smart money tracking, money flow analytics, trending markets, AI recommendations, etc. I thought it was executed exceptionally well. But when I asked why they ultimately pivoted to Perps, they gave me several highly logical reasons:

First, prediction markets cannot capture the dividend of spilling liquidity. Why was DeFi Summer so booming back then? A major reason was the US zero-interest-rate environment and massive liquidity injection. Liquidity spilled from US stocks into crypto, and from Bitcoin into altcoins, perfectly timing the DeFi summer explosion, elevating the entire industry. However, prediction markets inherently struggle to benefit from this spillage because it's a vehicle for expressing opinions, not an asset class.

Second, the frequency of attention isn't sufficient. Sports and politics only deliver major events every few days or even weeks. But Memes differ. Look at why pump.fun is so dominant—you scroll Twitter daily and catch new hot topics and memes instantly. Like the PNUT squirrel coin back then, this happens daily. Meme token issuances occur on a daily, even hourly cadence.

Third, lacking the ceiling effect of a "leading car." In Meme markets or traditional trading, many users follow trades, creating a trailing herd mentality with a leading car effect. The quintessential example is Elon Musk shouting out Dogecoin, pushing its market cap to $70 billion. Some rode Dogecoin to 100x or even 1000x gains, boasting an incredibly high or even infinite ceiling. But prediction markets lack this ceiling. At best, you push a market probability from 10% to 100%, yielding a maximum 10x return. Furthermore, chances are the instant you place your bet, the market odds are already priced in. Your edge from mispricing might only be 20-30%, so doubling your money already counts as a massive win.

Fourth, rational decision-making accounts for too large a proportion. When you place a prediction market bet, you must genuinely believe your IQ, information, or understanding surpasses others'. I might place bets on Apple, AI semiconductors, or Intel's yield rates, but I won't bet on an ice hockey or football match I've never heard of. You'll find each person can only analyze five to ten targets at most—countable on two hands. But Memes aren't like that. A meme trend reaches the entire Twitter feed instantly, and a single token address can rapidly accumulate from ten thousand to hundreds of thousands, even millions of holders.

Therefore, I find all these points highly logical, which leads to my reduced bullishness on prediction markets compared to last year: I acknowledge its PMF genuinely exists, but its ceiling is drastically lower than that of Perps.

Victor: How do you view projects combining prediction markets with Memes? Like 42space, which you invested in, incorporates bonding curves and certain Meme elements.

Lao Bai: Honestly, I was previously strongly against Memes and never really liked them. But after observing the prediction market model, I've become less bearish on Memes rationally speaking. I believe Memes represent an indestructible, enduringly vibrant format. To summarize, the biggest difference between the Crypto market and traditional financial markets is that we harbor far more gamblers here. Moreover, we provide tools robust enough for serious betting. Crypto is also the highest-volatility market globally, representing its inherent industry advantage. Meanwhile, many functions prediction markets serve in traditional finance can actually be substituted by options or Credit Default Swaps (CDS). Lacking the optionality premium from TradFi, and insufficiently gambling-heavy to satisfy degens in Crypto, prediction markets fall into a middle ground that satisfies neither perfectly.

My rationale for investing in 42space is twofold: First, I have a personal relationship with the founder; Second, although I lack the confidence to claim 42space will inevitably triumph, I believe the path of "combining Memes with prediction markets to make the prediction market itself more degen-friendly" is a worthwhile trajectory to explore.

XIII. Robinhood's Clear Strategy: RWAs for Investors, Memes for Degens

Victor: How do you view the on-chain Meme frenzy surrounding Robinhood? Recently, the original native launchpad on Robinhood shut down, and Uniswap newly launched the pools.trade launchpad on Robinhood Chain a few days ago. Although Robinhood primarily focuses on RWAs, a substantial portion of its trading volume is driven by Memes. How do you perceive the growth dynamic between this chain and Memes?

Lao Bai: I think Robinhood's strategy is exceptionally clear and brilliant. To be explicit, their chain does two things: RWAs and Memes. They intuitively understand exactly what users need. If you're a long-term investor, you need RWA assets like Micron, NVIDIA, or SK Hynix; you don't need worthless VC-backed vaporware alts. For the majority of degens, however, you crave high-volatility assets. So they hand you Memes.

When the Base chain initially launched, it also used Memes to drive initial traffic because Memes exceptionally easily trigger early 100x or 1000x pumps, acting as a massive natural attractor for attention. Robinhood masters this tactic perfectly: bring users in first, accumulate deposits. You might shift gears anytime, saying "I made profits on Memes and want to hold onto the cash, what do I do?" Buy more NVIDIA on Robinhood. Their maneuver explicitly clarifies what both crypto natives and TradFi circles are thinking and playing with.

XIV. Exchange Exits and Talent Flow to AI: Why This Cycle Is Exceptionally Severe

Victor: Recently, many exchanges announced halting trading operations, like formerly dominant BitMEX (inventor of perpetual contracts) and BitMart. Many projects also announced shutdowns, such as last cycle's prominent Zapper and Fantasy Top. Having survived multiple cycles, does this happen at the bottom of every cycle? Is exchange attrition particularly pronounced this cycle?

Lao Bai: Actually, similar events occur at the trough of every cycle, but this time is extraordinarily severe. In previous cycle bottoms, legacy exchanges like BitMEX, barring catastrophic security breaches, rarely shut down; projects like Zapper, which I personally used, favored, and recognized for genuine users and intrinsic value during DeFi Summer, typically survived past cycle nadirs. But now you see they simply cannot endure. We are genuinely entering a mature phase where the prior illusion that "growth masks all problems" has vanished.

Back then, profitability worked, or even if unprofitable, positive user/capital inflows ensured mutual harmony. Now, there are no fresh funds, nor user increments. Conversely, our talent and users are fleeing outward. Every trader and KOL discusses US stocks daily. Stripped of growth, all underlying issues surface simultaneously. Our ecosystem's shrinking circle means exchanges and projects that previously endured resilience thresholds can no longer withstand the pressure.

Victor: Talent exodus is indeed a massive problem, evidenced recently by Sui co-founder announcing a move to Anthropic. Contrasting with the 2017 and 2020 cycles when the world's brightest minds seemingly flocked toward Crypto, the dynamic has inverted. What phenomena have you observed locally? When might this trend reverse?

Lao Bai: I doubt we'll witness a reversal in the coming years; currently, I see absolutely no reversibility in this talent drain. My previous extreme bullishness on blockchain stemmed largely from the caliber of talent. I even tweeted in 2023: At ABCDE, we recruited four or five interns sequentially—three from Tsinghua, one from Peking University. Among colleagues, Siyuan holds a database PhD from HKUST, Joy hails from UPenn. During my two-plus years at ABCDE, I screened 1,300 to 1,500 projects mentally reviewing founders, roughly half graduating from Ivy League schools. I interviewed so many Harvard, Stanford, Berkeley, MIT founders. As a mediocre student from a provincial university, chatting with hundreds of world-class founders within one industry was unimaginable elsewhere, prompting my conviction: How could one NOT be bullish on Crypto?

But that was 2023, right when AI and ChatGPT debuted, before the disparity became stark. Look now; many acquaintances, including former interns and colleagues, migrated to AI. Aside from profit incentives, I believe the root cause is: Over the past five to six years post-Ethereum birth, we had fascinating puzzles to solve—distributed consensus, open finance, on-chain trading, MEV, scalability, censorship resistance, and privacy. However, by the 2024 inflection point, nearly all these foundational issues are adequately resolved. Emerging projects from 2024 onward, whether new chains or L2s, deploy nearly identical blueprints: launch a Meme for traffic, build a wallet, design tokenomics incentives, set up vaults/lending pools, fork Uniswap and AAVE. Everyone plays the exact same game, rendering development unsexy and unrewarding. Asking Harvard/Stanford/MIT elites to construct clones yields neither profit nor intellectual stimulation. Conversely, within AI, we possess countless unresolved challenges over the next five to ten years, mirroring blockchain's past decade. Consequently, from a problem-solving standpoint, I perceive zero probability of large-scale AI talent migrating back to Crypto.

Victor: Does Ethereum's innovation roadmap factor in? How do you assess Ethereum's developmental trajectory over the past few years? Back then, numerous brilliant minds entered with utopian ideals, yet those grandest visions appeared falsified, prompting departures.

Lao Bai: In hindsight, Ethereum executed remarkably well over the past few years. Many DeFi Summer innovations originated on Ethereum, encompassing the Layer 2 paradigm and the foundational smart contract architecture—all forged independently by Ethereum. Solana stands alone as the sole alternative approaching viability from a divergent architectural perspective. Ethereum effectively laid the bedrock. Minor refinements remain regarding privacy, staking yields, decentralization metrics, and ZK-proof implementations, yet these constitute peripheral details rather than existential flaws.

Over the past three to five years, we replicated and stress-tested every Wall Street mechanical strategy imaginable on-chain—from tranches and Pendle-like risk/return splittings to intricate CDS derivatives. Ultimately, functionality narrows down to essentially a DEX and a lending protocol. Overly complex architectures fail to gain traction on-chain. Pendle represented peak innovation by transplanting TradFi mechanics precisely to the threshold of user tolerance; attempting further complexity via GammaSwap, hedging instruments, options, insurance, or CDS yielded nothing. Even fundamental options protocols—having audited over ten on-chain iterations—remain nonexistent competitively, leaving centralized venues like Deribit unchallenged. Having exhausted viable on-chain innovation vectors, fewer urgent technical mysteries warrant global genius intervention. Cognitively, exploring AI presents substantially greater scientific allure.

XV. From Crypto to US Stock Options: Three Cognitive Upgrades to the Trading Framework

Victor: You've significantly allocated capital into AI and semiconductor equities this year while actively trading options. Transitioning from Crypto to equity markets, how have your trading frameworks evolved, and what key insights emerged? Options participation remained niche within Crypto historically; how do you leverage them for directional expression?

Lao Bai: Upon transitioning to equities, three aspects profoundly impacted my methodology:

Firstly, attempting to construct a robust systematic framework. Historically, my trading lacked systemic discipline, operating predominantly on narrative momentum. Post-DeFi Summer profitability engendered path dependency, driving oversized allocations into novel alts perceived as creatively superior. Nevertheless, the 2022 Luna collapse and late 2024 GOAT volatility eroded considerable gains despite avoiding leverage or derivatives. Substantial spot alts depreciation underscored immature structural methodologies—too young, too naive. Investing fundamentally sound US equities possessing authentic PMF, addressable markets, and active user bases diverges radically from Crypto paradigms reliant on speculative multiples, narrative extrapolations, and aesthetic novelty. Rigorous systematic architectures incorporating drawdown parameters, capital deployment calculus, and opportunity costs become mandatory.

Secondly, learning epistemic humility. Crypto harbors perilous cognitive habits among primary investors: finishing a whitepaper, auditing DefiLlama metrics, interviewing founders for hours, deconstructing token emissions, then concluding thorough comprehension precedes capital deployment—an illusion. Equity markets expose granular complexity: corporate earnings trajectories, margin profiles, inventory turnover, capex requirements, macro rate environments, derivatives positioning, implied volatility surfaces, supply chain dynamics, competitor landscapes. Exhaustive data assimilation remains structurally impossible. Acknowledging universal information asymmetry transcends mere rhetoric; recognizing omnipotence absence defines mature market cognition.

Third—and paramount—expanded expressive modalities. Crypto utilizes rudimentary binary mechanisms: bullish implies direct asset acquisition, elevated conviction triggers perpetual leverage amplification, yielding disproportionate rewards or ruinous liquidations. Traditional finance provides sophisticated stratagems: Nvidia overvaluation concerns manifest as covered puts; Samsung/Micron longevity convictions amid near-term volatility anxieties resolve through call spreads; forecasting SpaceX secondary launch volatility necessitated straddle configurations despite strategic failure, capturing directional neutrality exclusively. Semiconductor/AI portfolio overheating prompts protective put acquisitions. Multidimensional expression forces neurological clarification distinguishing underlying appreciation from variance speculation versus temporal decay trading. Systematic clarity dictates instrument selection, establishing profound personal methodological transformation.

Victor: Discussing entities like SpaceX, units such as Unitree Robotics approach imminent IPO status without available equities or options, though Hyperliquid facilitates pre-market perpetuals. How would you navigate pre-IPO asset allocation?

Lao Bai: Frankly, I perceive structural opacity, consequently abstaining. Similar historical precedent existed with SpaceX pre-IPO derivatives exclusively residing on Hyperliquid; I deliberately avoided deployment. Pure short-duration sentiment speculation devoid of underlying anchor securities generates bifurcated directional exposure. Even substantively accurate projections face instantaneous whipsaw liquidation risks across dual positional spectrums. Consequently, pre-launch perpetual vehicles remain excluded from my operational mandate.

XVI. Strategic Guidance Per Stakeholder: Founders, Traders, VCs, and Exchanges Maintaining Table Position

Victor: Concluding this dialogue, request strategic guidance tailored to distinct participant archetypes navigating bear market consolidation while maintaining competitive positioning.

Lao Bai: Structured differentiation by participant category:

Founders operating in contemporary primary ecosystems confront fundamentally altered parameters relative to prior cycles. Conceptual novelty or direct TradFi mechanic transplantation no longer guarantees viability. Successful venture creation currently necessitates institutional alignment—particularly North American resource integration—authentic PMF validation, and demonstrable user acquisition trajectories. Eliminating cryptographic abstraction may occasionally optimize product utility; surrendering PMF principles solely for ecosystem retention proves strategically counterproductive. Aligning with Haseeb's observation, cryptographic infrastructure should organically permeate application architecture. Architectural prioritization must center on user segmentation and functional utility delivery rather than mandatory token issuance or algorithmic governance imposition. Neither mechanism constitutes structural prerequisites.

Traders must suppress geographic or asset-class identity fixation. Classifications encompassing cryptocurrency, US, Taiwanese, Korean, or mainland equity portfolios remain operationally irrelevant. Absence of localized volatility mandates cross-market exploration directing risk capital allocation toward asymmetric payoff structures utilizing informational arbitrage or structural advantage.

Fellow venture practitioners occupy limited advisory authority given organizational closure trajectories. Extractive lessons dictate hybrid primary-secondary deployment strategies. Surviving contemporary funds demonstrate integrated capital allocation frameworks. Primary deployments incorporate pure equity instruments circumventing mandatory native token acquisition protocols. Secondary allocations facilitate strategic BTC/ETH accumulation or perpetual derivative hedging architectures alongside equities brokerage exposure or traditional marketplace participation. Abandoning artificial categorical confinement preserves flexible institutional positioning.

Centralized trading venues confront unprecedented existential recalibration requirements. Operational definitions transitioning beyond legacy digital asset facilitators necessitate competitive benchmarking against Robinhood, Interactive Brokers, Tiger Securities, or traditional banking conglomerates. Customer retention paradigms prioritize aggregate asset depth, liquidity density, fee compression, and interface convenience. Institutional evolution mandates transforming listing-centric models into universal global risk asset distribution architectures.

Prevailing architectural projection identifies Robinhood as definitive terminal exchange configuration. Unified ecosystem deployment spanning digital currency settlements, US equity purchases, crypto asset acquisition, leverage utilization, derivative execution, and predictive market participation establishes comprehensive service aggregation. Functional equivalency mirrors mobile ecosystem dominance paradigms. Executive enthusiasm regarding consolidated super-application architectures validates cross-industry convergence trajectories. Infrastructure integrations pursuing native payment network embeddings further corroborate unified transactional universes.

Victor & Mr. Z: Express profound gratitude to guest speaker for participating in broadcast programming delivering comprehensive analytical perspectives spanning cryptographic historiography, prospective sector trajectories, and artificial intelligence market surveillance. Acknowledge audience engagement throughout recording duration. Subscription facilitation encourages distribution across designated social networking platforms, newsletter aggregators, and video hosting services promoting thematic interest community expansion. Conclude session previewing subsequent episode publication.