Token2049 has ended, and Crypto entrepreneurs and practitioners from all over the world have left one after another, and the Crypto atmosphere in Singapore has returned to calm. After the intensive exchanges and discussions during this event, I have some new thoughts on the narrative of the new cycle.

The overall structure of the Crypto industry has been improved this year:

1. The most popular story of Layer0 is still Omnichain

Layerzero not only defines the full-chain interoperability of the ultra-light node mode, but also defines the new standard of Omnichain Fungible Token (OFT). OFT provides a more user-friendly and efficient cross-chain transfer solution. OFT has two versions, v1 and v2. The main difference between the two versions is that v2 can support non-EVM chains (such as Aptos, etc.). In addition, OFNT721 and OFNT1155 also support Omnichain NFT.

Many projects have adopted Layerzero's OFT standard, such as Ethereum layer2 Metis, Multichain stablecoin MIM, CDP protocol TapiocaDAO, Wrapped Bitcoin protocol BTC.b, multi-chain lending protocol Radiant.capital, etc.

For the Token2049 event, Layerzero dispatched a fleet of electric vehicles to safely transport attendees to any Token 2049 related activities on many streets in Singapore. Attendees only need to show their conference pass or wristband to get a free ride to the event destination and receive a small gift from Layerzero.

2. Layer 1 is no longer the first-tier narrative

In this cycle, people have different opinions on Layer1 projects. The new public chain of Move series has received a little more attention. Other public chains, whether Focus on DeFi or NFT, or public chains that support multiple programming languages ​​and can attract tens of millions of Web2 developers, do not seem to have gained broad consensus. (This only represents the communication experience during Token2049. Even if there is no broad consensus, it does not mean that the prospects are not good. Disagreement may lead to Alpha benefits.)

Many people no longer believe that there will be a new public chain that can surpass Ethereum. Even if they still believe that Ethereum may not be the end game, they are not willing to bet heavily on New Layer1. Of course, there are still many institutions that are still investing in New Layer1. For example, Jump Crypto, Coinbase, Circle Ventures, etc. are still common VCs in New Layer1 project financing information.

3. The Layer2 structure is basically set, and the Layer2 Middleware is also ready

There are mainly two strong Optimistic Rollups - Optimism and Arbitrum. The first-tier ZK Rollup/zkEVM mainly includes StarkNet, zkSync, Polygon, Scroll, Taiko, etc.

The middleware for ZK Rollup/zkVM is also in place. According to the Transaction Flow of ZK Rollup:

The key Middleware mainly includes:

  • Sequencer——The industry generally calls for the Decentralized Sequencer Network, and the leading projects include Espresso system, Astria, Altlayer, Radius, Madara, Fairblock, etc.

  • Prover——There are already quite a few projects doing ZK Proof. According to their expertise or well-known specific ZK Proof directions, they include: Nil Foundation doing Layer2 ZK Proof, Axiom focusing more on Applications ZK Proof, Risczero doing both Layer2 ZK Proof for its own Layer2 Bonsai Network and Applications ZK Proof, etc.; there are also Provers belonging to a certain project, such as Stone, Platinum, Sandstorm, etc. focusing on StarkNet CairoVM.

  • Hardware Accelerators - One of the constraints limiting the performance of ZK Rollup/zkEVM is that the zkp generation speed is not fast enough, so hardware acceleration projects for generating zkp have emerged, such as Ingonyama, which has GPU, FPGA and ASIC hardware acceleration, Cysic, which has launched SolarMSM (based on FPGA to accelerate MSM computing performance), and Accseal, which is deeply engaged in privacy computing hardware acceleration.

4. There is no shortage of cash cows in Middleware

After ETH2.0, the LSD track has become the second "cash cow" business direction of the Crypto industry after Bitcoin PoW mining. Under the Ethereum PoS mechanism, the annual output value of Staking is about 680,000 ETH (over 1 billion US dollars). There are also some very high-quality or potential projects, such as Lido with the highest market share, Alluvial mainly targeting B-side users, and Puffer that can prevent Slash.

Since Flashbots proposed MEV-Boost as a practice of PBS outside the protocol, the MEV track has already begun to preview the future profit distribution mechanism before Danksharding was implemented. For MEV value capturers, in order to obtain more MEV opportunities, there must be sufficient Orderflow; for users, in order to avoid their transactions being sniped by MEV value capturers, it seems that they can only choose to join if they can't beat them. Because there is no way to completely eliminate MEV, this business must exist. At present, the more reasonable way to exist is to feed back MEV income to the User's Wallet, DEX or other dapps.

5. Gao Qi’s “risk-free return” creates application layer opportunities for LSDFi and RWA

ETH Staking income has become the "Treasury bond income" of Crypto, and LSDFi further helps to increase the yield of ETH holders. According to the current Lido data, the annualized yield of Ethereum's PoS is about 3.6%, which is similar to the positioning of national bonds in the Crypto industry. After the Shanghai upgrade, ETH Staking can basically be withdrawn at any time, and there is sufficient liquidity between stETH and ETH to exchange at any time. This greatly increases the confidence of long-term ETH holders to participate in Staking to obtain ETH2.0 mining income.

With stETH and other assets, ETH, which is equivalent to Staking, has regained liquidity, which has led to the emergence of more LSDFi applications. For example: Lybra Finance, a stablecoin supported by LSD, has an annualized rate of return of about 7% to its holders due to its underlying pledged assets being interest-bearing assets in the form of stETH, plus token rewards, making its stablecoin eUSD an "automatic interest-bearing stablecoin". The TVL of the Lybra Finance protocol is currently about $72M; Pendle, an interest rate swap protocol, divides the income tokens into principal tokens and income tokens, giving users the opportunity to execute various income management strategies in it. The TVL of Pendle Finance is currently about $144M; there are also some DeFi protocols that have risen because they have captured more LSD Tokens, such as the emerging AMM Maverick, which has successfully become a first-tier AMM due to its high market share of LSD Tokens such as swETH, wstETH, and cbETH.

The annualized yield of ETH Staking of about 3.6% plus the additional income of LSDFi can indeed allow long-term ETH holders to capture more income. In contrast, the annualized yield of US Treasury bonds of about 5% has pushed RWA to the forefront, but this can only be said to allow existing Crypto users to enjoy the higher yield of US Treasury bonds in recent times, but it cannot actually prove that RWA can help Crypto attract new users. In short, the "risk-free rate" at the current stage is indeed a bit high, and for long-term Crypto holders, it is a little star in the bear market.

6. With Mass Adoption as the goal, lowering the threshold of user experience is the underlying demand

There are many innovations in the User Service Layer of this cycle. The common narrative is to lower the threshold of user experience. The ultimate goal is the same, which is to achieve Mass Adoption.

Intent Centric is a hot topic recently, centered on intent. Several leading projects have different entry points. Bob the Solver focuses on attracting more users, dappOS aims to lower the user threshold, Anoma aims to help users express their needs better, and Essential focuses on helping users reduce MEV squeeze. In fact, in summary, these projects are either solving the problem of "difficult to use protocols and more difficult to combine protocols" or solving the problem of "maximizing user benefits".

TGBot is another hot topic and one of the practical applications of Web3+AI. Telegram bots are small applications that run in Telegram. Users interact (talk) with the robot in the chat window of Telegram. They only need to use some simple commands to support users to monitor token information, do some on-chain DEX transactions, copy transactions, etc. With the increase in the number of users, transaction volume and token prices of Unibot, TGBots quickly gained market attention. In a very short period of time, TGBots has developed trading, analysis, airdrop & mining products. In essence, TGBots actually lowers the threshold of user experience and helps to achieve Mass Adoption.

7. This time, Token2049 has many projects combining Crypto and AI

The combination of Crypto and AI is the choice of the times. Crypto and ZK technology create Trustless, non-sovereign currency and smart contracts are more suitable for intelligent robots. There are many entrepreneurial projects combining Crypto and AI in this Token2049 event. After the communication, I was a little shocked and touched.

Carbon-based humans created silicon-based AI to better achieve automation, improve production efficiency, process large-scale data, and explore the wider universe. In this process, AI was gradually given the ability to think. The Earth, which is 70% ocean and has oxygen, is suitable for carbon-based life; but in the wider oxygen-free and water-deficient universe, silicon-based life is easier to survive than carbon-based life. Crypto, aerospace technology, and green energy developed by humans may eventually be used in AI, allowing intelligent robots to better survive and produce in the universe, thereby better serving humans.

At present, I have seen the following combinations of Crypto and AI:

  • On-chain interaction automation: TGBots represented by Unibot. For many new users, the threshold for using Web3 on-chain products is too high and difficult to use, and Bots products have a market. Telegram has more than 500 million monthly active users and is relatively open and Crypto-friendly. The emergence and popularity of TGBots is a natural result.

  • To Earn empowers AI business: Public.ai is a representative example. Public.ai is a labeling company focusing on the medical field. On the one hand, it provides professional medical labeling tools and labeling services. On the other hand, it integrates global professional labeling talents in a decentralized manner to improve labeling quality and reduce costs.

  • Web3 version of ChatGPT: represented by Lasso and Kaito. Lasso is a natural language search engine for on-chain data. No SQL or other complex smart contracts are required. You can search chains, data, etc. using natural language. For example, if you want to know the TVL of Lido and Rocket pool over time, users only need to enter the natural language "Show TVLs of Lido Finance vs Rocket pool over time", and Lasso can present the results in the form of charts.

  • ZKML: Modulus Labs is more vocal in this direction. With ZKML, we are talking about creating zero-knowledge proofs for the reasoning step of ML models, not ML model training. The effect of ZKML is that the prover can prove the computational correctness of the ML model without revealing any further information. Modulus Labs is building some use cases, including RockyBot, Lyra Finance, Self-improve blockchain, etc.

at last

Singapore is an important hub for the Web3 industry to connect the East and the West. The Token2049 event had over 10,000 attendees, and is said to be the largest event so far this year. Most of the important projects came, with service projects participating in the main booth and narrative projects hosting side events.

I visited all the booths at the main venue of Token2049 and learned about the approximate costs of various booths. I found that the exhibition was mainly attended by public chains, Layer2 projects, exchanges and payment solution companies. Perhaps the exhibition format is more valuable to the business of such companies.

Side Events are mostly hosted by leading Infra projects or VCs, and are mostly co-hosted by Infra projects’ ecosystem projects and VC portfolios. In terms of event formats, there are about half social Dinner Party and half competitive Hackathon/Workshop + Demo Day.

Still need to continue building.