DeFi notes to girlfriend: I rely on time for insurance

BTC0.0₂₀

ETH0.0₂₀

HTX0.0₂₀

SOL0.0₂₀

BNB0.0₂₀

BTC0.0₂₀

ETH0.0₂₀

HTX0.0₂₀

SOL0.0₂₀

BNB0.0₂₀

DeFi notes to girlfriend: I rely on time for insurance

Kay

2020-10-17 12:03

本文约11002字，阅读全文需要约44分钟

A person who is strong enough can get used to the losses he can accept. If the insurance is really falsified, it is also an option for socialized losses to become a norm.

Author: Kay, a male student without a story.

This is the second article in the "DeFi Notes for Girlfriend" series. This series is dedicated to providing some common sense about DeFi in the most easy-to-understand language, including not limited to conventional concepts such as impermanent loss/AMM, or Regarding the analysis of the principle and design of the new project, we strive to make it possible for girlfriends (imaginary and possible) who do not understand DeFi at all to understand. I hope I can write at least 30 articles.

Table of contents

Disclaimer

I don't have a girlfriend right now
I love women
The author is extremely opposed to gender discrimination that has no scientific basis, such as "boys are born to study science, and girls are better off with natural arts"
The author is not a professional financial person. This is a non-serious series. People who read it can read it casually, and those who write it can write whatever they want.

Retro

DeFi Notes to Girlfriend: Impermanent Loss

Aspects of Classical DeFi Insurance (Before Cronje)

The time has come to 2020Q4. Recently, the daily gas idle time has been maintained at around 30Gwei. At present, most DeFi players regard Q4 as the garbage time of the agricultural holiday. It is always difficult to reverse the trend. If we want to give DeFi 2020 If Hype chooses a person of the year, there is a high probability that no one will dare to choose Andre Cronje (YFI developer, often abbreviated as AC in Chinese circles).

When YFI’s current price was in full swing, someone on CT (Crypto Twitter) even redefined BC (Before Christ) / AC (AD):

Before Cronje (before YFI appeared)
After Cronje (after the appearance of YFI)

The follow-up Chad and YOLO frenzy caused by YFI will not be repeated here, see the picture:

The point here is that, whether you like him or not, or whether you have been recognized by AC's follow-up platform/developed projects such as SAFE, EMN, LBI, etc., it must be admitted that AC is indeed a good at both economic model design and development There are not many geniuses like this. At the same time, YFI may really open aThe age of fair launch projects。

Classification

When it comes to DeFi insurance, most people’s first reaction is of course Nexus Mutual (NXM). “Mutual” in Chinese is generally translated as “mutual insurance”. This concept, the most familiar case for Chinese people may be Alipay’s Mutual Insurance. The first thing to say here is that “mutual insurance” is a very rudimentary form of insurance, which is still somewhat different from real modern insurance. Here, mutual Take Bao as an example, extract a comparison on the Internet:

1. Differences in insurance contracts

For modern insurance, when purchasing an insurance product, an insurance contract will be signed with the insurance company; the contract clearly stipulates the guarantee responsibility, compensation standard, liability exemption and other contents.

Joining "Xu Hu Bao" does not sign any contract, and you can only learn about the protection content, compensation standards, etc. through the platform announcement.

2. Differences in payment of premiums

When purchasing a certain insurance product, modern insurance companies will clearly inform the premium rate that the premium will not change in the next few decades.

The premiums of each phase of Xiang Hu Bao are different, and each phase is capped at 188 yuan, that is, the premium of each phase is unknown before reaching the cap.

3. When there is a dispute over the claim, the handling method is different

Insurance: You can apply for arbitration or prosecution, and use legal weapons to protect our legitimate interests.

Xiangxiangbao: Started the jury (like DAO), on March 26, 2019, the first compensation trial case was launched, and 900,000 compensation judges quarreled. The 900,000 jurors are neither legal nor medical. But it depends on these unprofessionals to decide whether to get compensation.

In addition to mutual insurance such as NXM, the current DeFi insurance business in a broad sense also includes:

Financial derivatives (such as options), such as Opyn
Prediction markets like Augur
CDS, social network guarantees such as CDx, VouchForMe
image description

Image source: Cointelegraph China

Yes, the three major factors affecting the value of an option are the underlying price, volatility, and time. There are also three factors that affect the value of insurance, the price of the insured asset, the duration of insurance, and the risk of the insured asset (that is, the volatility of the insured asset). Volatility is risk, and its academic definition is the standard deviation of the rate of return, which is used to measure the risk of assets.

Therefore, it can be considered that the essence of options is also a kind of insurance in a broad sense. Models such as prediction markets and CDS are similar.

In general, the current DeFi insurance is basically based on crypto-native business, and has not yet involved large-scale off-chain business, and the number and types are relatively limited. Of course, it can be considered that there is still a lot of room for development. A few typical projects are introduced below.

Nexus Mutual

In this year’s DeFi Hype, Nexus Mutual is almost the only head project of mutual insurance in a long period of time. Nexus Mutual is a community-based insurance platform. Token holders are both income earners and risk takers. At the same time, the tokens they hold are used to determine how much insurance a certain project can have, and to decide whether to approve the claim.

Of course, this is also a point of criticism. An inaccurate analogy is - let the shareholders of the insurance company decide whether the compensation is worthy, of course, if you can’t pay, you won’t pay.

The current NXM payment resolution model is:

The first round: voting among the claim assessors, if more than 70% of the people reach a consensus, the result will be issued directly, if less than 70%, the second round will be carried out. (If the minimum number of votes is not reached, all members will participate in the vote)
The second round: All Nexus Mutual members participate in voting, and the consensus of most people is passed directly. (If the minimum number of votes does not reach the result of the first round of majority selection)

In fact, so far, the only claims passed by Nexus Mutual are three flashloan attacks suffered by bZx in February this year, and more than 20 other claims have been rejected by community governance. This also includes, for example, when MakerDAO plummeted on 3.12 this year, due to network congestion and liquidation mechanism problems, some ETH was liquidated at $0, causing users more than $5 million in losses. These losses were not covered by insurance (although MakerDAO Subsequently, the user was compensated through the auction of MKR). In other words, currently NXM only accepts compensation for contract security layer issues, and it seems that NXM cannot achieve the purpose of transferring risks well for losses caused by mechanism issues.

You can find all projects that support insurance on the Staking page (currently supports 39 DeFi products). Most NXM is concentrated on the insurance policies of head projects such as Compound, Curve, and Aave. This is of course because users are more secure about these projects. Not recognized, on the other hand, it is also because these projects have a large amount of purchase, and the insurability of some tail projects is very small, even only a few Stakers and hundreds of NXM.

Nevertheless, according to DeBank data, since 312, the total locked volume (TVL) of Nexus Mutual has reached 82.54 million US dollars, an increase of more than 3000%. This is of course related to this round of DeFi Hype, but it can also be speculated from the side ——"If the compensation is up to the 'shareholders', then users who are underwritten by 'rogues' can rest assured."

In addition, the YFI-based product yinsure.finance mentioned later is currently connected to NXM as an insurance underwriter, so NXM may be one of the few projects that has not been siphoned by YFI in the summer, but has enjoyed dividends.

In addition, I have to mention the model of NXM token itself, which uses the bonding curve IBCO (Initial Bonding Curve Offering). The characteristic of this model is that because the price is determined by a fixed function, the amount of tokens generated is No matter how much the price is, it will almost inevitably rise sharply in the early stage, and it is relatively difficult to skyrocket after the initial stage. On the other hand, because the token used for mint is ETH, the price correlation with ETH is strong. From the price point of view, it is more like a leveraged ETH at present.

The tokens that later adopted the IBCO model obviously noticed the latter point. For example, the NFT concept coin GHST of the AAVE ecology directly used DAI to mint to reduce the price correlation with ETH.

In addition, the use of the entire NXM platform and the mint of the token itself require KYC, but someone has created WNXM (an ERC-20 token anchored to NXM). Currently, both Huobi and Binance are WNXM, because mint and destruction The behavior of NXM will also bring about price fluctuations, and there is often an arbitrage space here.

Opyn & Hegic & Opium

Opyn

Opyn is a 19-year-old project. At first, it wanted to be a margin trading platform like the current dYdX. It once provided the highest leverage available for ETH in the DeFi world (6 times, while dYdX has only been around 5 times for a long time). Later, the project pivoted to Compound Insurance, and then pivot to options on the chain, currently provides American options of ETH, UNI, WBTC, SNX, and YFI (different from European options, American options can be exercised at any time), but the liquidity is average, which also leads to relatively small products. (another reason is that option products are mainly created by the team), for example, YFI only has a put option with a strike price of 20,000 USDC.

Once the option is created, the user will mint the corresponding oToken certificate (ERC-20 token), these oTokens can be traded on Uniswap, and option sellers can sell these oTokens on Uniswap.

In addition, Opyn also provides insurance services similar to NXM, but currently there is only one product, Compound principal insurance.

What I have to mention here is that Opyn’s ETH smart contract suffered a hacker attack in August, causing a loss of 370,000 US dollars. It is said that the relevant contract was not audited at the time of the incident. A similar white hat attack took out the remaining pledged funds in the vault 😂 and promised to pay for related losses.

Opyn is also developing the V2 version, which may add new functions such as margin, multi-currency mortgage, and trading system upgrades.

Hegic

Hegic is roughly similar to Opyn, and it is also an option protocol on the chain. It currently supports ETH and WBTC. The difference is that the liquidity providers in the Hegic option model are the collective sellers of options, but they will not directly create options for each option. Customized by the buyer, all option products can be selected by the buyer on their own, the type, strike price, and term, which is very flexible.

The option price is pre-determined by the Hegic algorithm. The option seller can automatically obtain the relevant option price by entering the option terms that he intends to buy. At the same time, Hegic will manually update the most important component of the option pricing in the Black-Scholes formula based on historical data—— Implied volatility. Hegic's option prices are derived from data published by skew.com, and then the implied volatility (IV) is manually updated in the Hegic protocol instead of being dynamically priced by the market in real time. Therefore, arbitrage opportunities may arise between Hegic and other options platforms.

My definition of Hegic is "on-chain option vending machine", or the Uniswap (AMM) form of options, but according to Uniswap's experience, it seems that this model is more suitable for long-tail assets, and those who need options tools generally It is an asset that has become a "popular lover". I have to say that the overall idea and implementation are good, with the simplicity that men of science and technology like, and like AMM, it is something that has never existed in traditional finance. But whether the foundation of existence will be falsified is another matter.

Like Opyn, despite being audited, Hegic has historically had two significant vulnerabilities.

Opium

Opium means opium. As the name suggests, although it uses the traditional order book model, it is a real fucking car. The options supported include very aggressive binary options, except for the traditional "popular lover" coins. In addition to options, there are some innovative products, such as:

Ethereum fee gas options to help farmers hedge the risk of gas fluctuations
The COMP option OEX-ZEPO-1*COMP before the launch of COMP, with an exercise price of 0, realizes a product similar to the traditional CEX "futures stand-alone currency", but the traditional CEX futures currency still anchors the real currency for future delivery in name Yes, and through options, it has nothing to do with the actual delivery currency, and can realize the transaction before COMP goes online for users

In addition, compared with the 100% margin requirements of Opyn and Hegic, Opium currently only requires a 33% margin requirement for seller users who use DAI, and supports multi-asset margin at the same time.

Etherisc

Etherisc can be traced back to 2016 and 2017, similar to most of the same period, with a big vision and awkward implementation. Many people say that he is the Aragon of the insurance industry on the chain. He wants to be a general decentralized insurance application platform, so that developers can use this platform to quickly develop new insurance products.

The Etherisc core team has developed some insurance common infrastructure, product templates, and insurance license-as-a-service (insurance license-as-a-service), allowing anyone to create their own insurance products.

At present, the Etherisc community has designed a set of basic insurance products, ranging from flight delay insurance, hurricane insurance to encrypted wallet and loan mortgage insurance, but most of them seem to be just demo tests.

A new story for DeFi insurance (After Cronje)

yinsure.finance

yinsure.finance is not DeFi + NFT, but CDS in NFT shell.

When looking back at articles about the 2008 financial crisis, some words such as CDS and MBS always pop up. Here is an explanation of CDS:

1. CDS is actually not a new thing in the true sense, it is a guarantee

2. Zhang San borrowed money from Li Si, afraid that Li Si would not pay it back, so he asked the "old uncle" who everyone trusted to guarantee

3. The financial brickman Ye Guantianxiang pressed the Monte Carlo method regression analysis on the calculator randomly, and told the old uncle that Li Si had a 2.56% probability of not repaying the loan

4. The old uncle charged Zhang San 10% of the guarantee fee, and signed a guarantee contract. If Li Si ran away and refused to pay back the money, the uncle would pay the full amount to Zhang San at his own expense.

So far, it is a conventional guarantee, which has been around for thousands of years, and is nothing new.

5. What if this contract can be bought and sold in the secondary market?

Li Si won the 5 million lottery, Zhang San is not worried that Li Si will not pay back the money
Zhang San felt that the guarantee fee for the old uncle was a bit of a loss, and he wanted to sell the guarantee contract at a price of 5%
Wang Wu, the mutual friend of Zhang San and Li Si, appeared. Wang Wu always felt that Li Si was unreliable and too ostentatious, and something would happen sooner or later, so he took over the guarantee contract
Li Si really took the 5 million won to speculate in the crude oil contract, not only lost money, but also owed 10 million

6. Ending

The money Zhang San borrowed was not recovered, and the sale contract continued to lose 5%.
Li Si became an old Lai
The old uncle returned the guaranteed loan in full to Wang Wu who held the guarantee contract.

What, what is CDS?

This guarantee contract is CDS, the essence of which is that it can be traded. Because it can be traded, even though people like "Wang Wu" did not actually lend money to Li Si, they could still make a profit through CDS, in this case, twenty times the profit.

The real prototype of the movie "The Big Short" has been suffering from the fact that there is no way to directly short the US housing market, because mortgage bonds have neither contracts nor options, until such products as CDS are discovered:

I don’t comment on whether the CDS model is good or bad here. From my personal point of view, if everyone knows the rules of the game, there is nothing wrong with it. When people are beaten, it depends on when people make money. The story is that Li Si paid back the money normally)

Those who think that these derivatives are bad, most of the time, it is just that human nature is risk-averse, and if they lose, they will turn the tables.

Okay, so much has been explained, back to yinsure.finance, the initial description of AC is like this, it looks like it is just an NXM that does not require KYC:

Insurer Vaults: The insurer deposits funds into the vault to form an insurance fund pool, and usually collects the weekly premium paid by the insured, and when the claim is approved, the fund pool is used to pay the insured;
Insured Vaults: The policyholder deposits funds into this vault and becomes the insured. The insured pays an initiation fee of 0.1% upon deposit and a weekly premium of 0.01%; can be withdrawn at any time;
Claim Governance: Insurance arbitration through community voting to determine whether a claim is required.

But when yinsure.finance was actually launched, everyone at Dadao found that things were not simple. Everyone said that Andre had made a combination of DeFi insurance and NFT.

If you understand it carefully, the concept of NFT here is actually not important, but NFT is convenient to distinguish the type of insurance, the duration and the amount of compensation, and it can be conveniently traded on a platform like opensea-this is a feature that NXM does not have, it can be traded Insurance policy, yinsure is a CDS in the cloak of NFT, of course, it is relatively rudimentary in form.

According to data from the Opensea platform, the transaction amount of yinsure.finance’s NFT policy on the platform has reached 5266.6 ETH in the first month, or about $1.96 million.

As I wrote this, I suddenly had an idea, since Hegic can be used as an "AMM" for options, will there be an "AMM" for CDS? (Single-digit transaction fees on platforms like opensea are still not low) I believe that similar products will appear soon.

If you cannot imagine the impact of CDS on the encrypted digital currency market now, you may wish to refer to the CDS market size in the traditional financial field. Although CDSs were only created in the 1990s, the market has grown rapidly. According to data from the U.S. Office of the Comptroller of the Currency, between 2000 and 2008, the size of the U.S. CDS market expanded substantially year by year, from $312.4 billion in the first quarter of 2005 to $1,589.7 billion in the fourth quarter of 2008. The proportion of the total derivatives market has remained above 85%, and it once became one of the important sources of income for US investment banks.

Another example is that in the modern foreign exchange market, 95% of the trading volume comes from speculation and 5% comes from "real demand".

yieldfarming.insure

In this DeFi Hype, in fact, for quite a long time, AC has almost been a religious image of gods. People's criticism of AC still starts with AC's retweet to yieldfarming.insure.

To put it simply, this is a project that AC sponsored $25,000 to develop (later AC mined and sold hundreds of thousands of dollars on this project), the token SAFE, perhaps readers have heard of it. Before the market turned bearish, the two core developers collapsed at the early stage of the project because of "division of labor and uneven distribution of spoils". Simply put, it was a "scumbag who cheated college students into working for millions, and internal strife and forks might return to zero. Everyone will eat melons." Watching a play" story.

Later, one of the college students said that he had dropped out of school for a year, and would continue to stick to his crypto ideals, love DeFi, love Andre, and would continue to develop this project, so there was a migration of SAFE2.

In terms of mode, due to the particularity of the encrypted digital currency market, the use of such CDS products is not limited to this. yieldfarming.insure (SAFE) introduces the insurance policy (yNFT) generated by yinsure into its own mining mechanism: pledge the yNFT of yinsure.finance into SAFE's official contract, and you can get SAFE token rewards. Supported by the insurance policy of yinsure.finance, the SAFE project was very popular at first, and its price once reached $4,000; however, due to internal disputes in the community, the currency price plummeted. Although the SAFE project ended in failure, it also shows the profound impact of CDS products on the cryptocurrency market.

"I rely on time" of DeFi insurance

This paragraph pays tribute to Mr. Wang Chuan's "I rely on time", but it is not like Mr. Wang Chuan's writing about the possibility of Tesla's success. It has already been written above. It may be possible to achieve a corner overtaking through the CDS mode. What I want to write here is The failure/falsification/alternative of DeFi insurance is possible.

single point of failure

In the Internet operation and maintenance industry, there is a very important concept "Single Point of Failure". Large companies generally have remote multi-active solutions for storing data, even if an earthquake/flood/debris flow in a certain place sweeps away the IDC computer room (Don’t laugh, earthquakes must be considered as normal disasters. Floods are okay for ordinary IDC computer rooms, but for BTC mines next to hydropower stations, it is a risk that must be considered). There may still be another computer room in a desert in the southern hemisphere. The same data, and if something goes wrong, you can switch services immediately.

Similarly, in the crypto world, there is also a multi-signature scheme for private key management, which will not be discussed in this article.

There are two types of single point of failure, recoverable and non-recoverable, for example:

Let's say you have a laptop for everyday use and the data on it is not backed up. One day, if the memory module of the notebook suddenly breaks down, the entire notebook will be unusable (unusable). But if you find after-sales maintenance personnel to help you change a memory module, then this notebook can be used again (recoverable).
Now change the assumption: it is not the memory module that is broken, but the hard disk is completely broken. At this time, the notebook is also in an unavailable state. But what's more serious is that even if you ask the after-sales maintenance personnel to replace a new hard disk for you, you can't get back the original data. This is the so-called "non-recoverable".
In these two examples: the single point of failure caused by the memory module is "recoverable", while the single point of failure caused by the hard disk is "unrecoverable" (unless you have a backup mechanism).

Obviously, non-recoverable single points of failure are more dangerous, and the current DeFi insurance model is more likely to encounter exactly the kind that is not recoverable. But without rushing to expand, let's look at two failures in the traditional IT world.

OpenSSL Heartbleed & Intel Meltdown

Heartbleed

Heartbleed is a well-known vulnerability in the first half of 2014. When this vulnerability was discovered, it had existed for more than two years. The Heartbleed vulnerability allowed hackers to read the confidential data in the system memory that should have been protected by the OpenSSL software (SSL is not accurate. The description is, those websites with links starting with https://, a tool for self-protection).

These protected data may include web content that should be encrypted, personal account passwords, and even credit card transaction-related data. Hackers can also use this loophole to monitor communication content on the network (such as instant messaging conversations).

The flaw affected millions of SSL certificates, including nearly every large website you could think of or heard of, and until a month and a half after the vulnerability was publicly disclosed, of the 800,000 most popular TLS-enabled websites, there were still 1.5% has not been fixed yet. The most surprising thing about Heartbleed at the time was not its influence. On the one hand, it existed for a long time. On the other hand, it exposed such an important infrastructure-level protocol as OpenSSL. Without enough donations, developers have been using love to generate electricity for a long time.

In that year, Mr. Luo Yonghao, a big V who has transformed into a live broadcast today, released the Smartisan T1, Smartisan’s first mobile phone. The 1 million yuan revenue from that conference was fully donated to the OpenSSL Foundation. Donated a lot, mainly from the conference ticket revenue. According to public information, the total donation of Hammer to the OpenSSL Foundation is more than 4 million RMB.

I'm not a fan of Luo Yonghao, but in this matter, respect.

Intel Meltdown

Intel Meltdown, a well-known vulnerability at the beginning of 2018, is a hardware-level vulnerability that is even more terrifying than software-level vulnerabilities such as Heartbleed. This flaw allows low-privilege processes to obtain memory protected by high-privilege regardless of whether they are privileged or not. Data in space, exploits are time-based side-channel attacks. It involves most of Intel's x86/x86-64 microprocessors, some IBM POWER architecture processors, and some ARM architecture processors. In other words, almost the entire small host and mainframe market, and the personal computer market have all been spared. .

In addition, the high degree of danger of the flaw (access to the storage space where sensitive data is located without privileges) once made information security personnel and organizations doubt the authenticity of the flaw, and the announcement of these flaws in advance is likely to trigger global information A security disaster, so this vulnerability was announced after the security personnel chose to contact the processor manufacturer and core customers to negotiate and prepare a repair plan.

At present, this vulnerability is only circumvented by software. Although Intel has repeatedly emphasized that the security fix will not significantly affect the performance of the processor, Microsoft's tests have shown that in the Windows environment where the vulnerability security fix is installed, the processors produced before 2015 The performance of Intel processors will drop, and the drop can even reach 30% in some tests. The fundamental fix for this vulnerability is to redesign the microarchitecture of the processor. For this reason, Intel, IBM, and ARM will all use new processor microarchitecture The launch schedule has been significantly delayed.

As an aside, here is what Alan Kay said in the 1970s:

People who're serious about software should make their own hardware. ——Alan Kay
Those who really care about their software will make their own hardware.

Alan Kay is also one of the people Jobs admired. In 1979, he invited Jobs to visit the Xerox Laboratory (Xerox Palo Alto Research Center) where he worked. In the 1980s, Apple brought Lisa and its successor to the world. (Graphical user interface) can appear in a real product, and the GUI is the Xerox Alto prototype that was stolen during the Xerox lab visit.

In November of the same year that Apple released Lisa, Microsoft officially released Windows 1.0, which is Microsoft's first operating system with a graphical interactive interface. Since then, the MS-DOS command line operating system has gradually come to an end, and the Windows graphical interface dynasty has begun. The reason why Microsoft switched from the command line interface to the graphical interface is precisely because of the influence of Apple Lisa.

Another influential statement by Alan Kay is that "the best way to foresee the future is to create the future yourself." There is no doubt that Jobs' Apple in the 2000s is the practical application of this philosophy.

DeFi insurance, how far is falsification?

Look back at the three points just mentioned:

Single point of failure and irreparable, the most terrible
The Heartbleed vulnerability exposed the security and insufficient incentives of a large number of open source projects and can only "use love to generate electricity"
Intel Meltdown vulnerability, lower-level architecture may also have problems

Didn't understand? Read it again.

Book that, all possible?

Yes, it's all possible.

Contracts on the Ethereum network generally cannot be soft-upgraded, and usually can only be hard-updated by changing the contract, and this operation generally needs to be done manually by the user, that is, it cannot be automatically repaired
Almost all DeFi insurance projects are open source projects, and insufficient incentive mechanisms are a common problem (of course, there are also well-designed ones)
The good news is that a large number of DeFi project developers have been freed in the last round, and the focus of participating in project development is self-realization rather than making money, but anyway, the presence or absence of incentives is still very different
Is it possible that there is a problem with the underlying architecture of DeFi? Too likely. For example, Synthetix's contract is recognized as a relatively well-implemented project, has also been audited, and has been reused in a large number of other projects (reusing audited code is of course safer than rewriting the probability), such as yearn. The finance (YFI) contract code even contains redundant code copied from the Synthetix contract (can programmers be called copying, not just reinventing the wheel), and various projects after YFI, such as YFII, YAM, Sushi, FARM, etc., naturally reused a lot of YFI codes again
Well, although the contracts of Synthetix, YFI, YFII, etc. have all been audited, is it possible that there is a subtle loophole like Meltdown?
If it really exists, will it cause a collective storm of DeFi projects because of the extremely high degree of code reuse? If this happens, how does DeFi insurance pay out?
Or, is there a potential loophole in the insurance contract itself?
Or, at a lower level, are there vulnerabilities in the Ethereum network itself that can be exploited?

……

The current DeFi insurance industry, especially NXM, is like a life insurance. All customers are grandpas who are around 60 years old and have just retired:

Limited first year payout
Limited payout for the second year
Limited third year payout

……

Until around the 20th year, Super Mario begins (Chinese male life expectancy is around 80 years old).

Of course, the current compensation logic is that the "insurance company's shareholders" decide whether to pay or not. Perhaps if this happens, it may not necessarily drain the insurance fund pool.

But by then, how many people will continue to believe in and support this model?

alternatives

There are currently two models: fork and treasury model.

fork

Ethereum Classic is the real Ethereum, no explanation (manual dog head

There are too many articles in this area, search for "The DAO" incident by yourself, there is no need to reinvent the wheel here.

treasury model

MakerDAO starts the mechanism of auctioning MKR in the case of insufficient collateral, refer to 312 Maker's solution to user losses.

accept loss

"I'm used to losses."

A quote from an old friend said that a person who is strong enough can get used to the losses he can accept. If the insurance is really falsified, it is also an option for socialized losses to become a norm.