sleepy0x13

Still have cookies? Increase position $KAITO

Realized that the memes we have been chasing for so long are meant to prepare us for this era.

Hashkey's Conflux moment

So funny, to receive the Humanity airdrop you first have to buy coins at the exchange as gas

I have never seen something so abstract

I have claimed 53 vibes this week! Claim yours at http://www.vibes.fun/?ref=sleepy0x13

Thanks to all the family members for your support. This week I secured the first place in @Mira_Network Yapper, and I would like to share briefly how I did it.

To be honest, I used to be quite laid-back about various rankings and never made a deliberate effort to climb the charts. I didn't really pay attention to points or anything, just played along with the flow. But this time is different; the Mira project itself really moved me, so I was more willing to invest some thought into it.

I have always believed that doing hardcore AI entrepreneurship in the Web3/Crypto field is unrealistic. Competing in computing power and large models, we can't outpace the big companies. However, if we can use some clever strategies and leverage existing resources, engaging in 'smart' AI projects might actually present opportunities. And Mira aligns quite well with the direction I've been looking for.

So I basically check Mira's official Twitter every day to see if there are any updates.

But the content I post is actually not much; I don't focus on quantity but rather on updates that I genuinely want to talk about or have ideas about. I write some lengthy content and express my viewpoints. It's not just a list of information; instead, I try to base it on my own understanding, and I have even pointed out its shortcomings in the past.

Pay more attention to project progress, engage in more independent thinking and viewpoint expression, and reduce the amount of homogenized information transfer. This might help improve Mindshare and also give you a clearer understanding and recognition of the project.

I've heard that many Crypto VCs are now turning to the US stock market, both primary and secondary, is that true?
Are there any VC experts here to discuss? 🥹

Circle stock is rising so sharply, USDC has reached $2

After learning that @0xinfini suddenly stopped its U Card business, I immediately contacted Princess @0xsexybanana to do an exclusive interview with her.

Yesterday, I talked with the Princess for an entire afternoon about why they wanted to create Infini, from the inception of the 'financial product' to later veering off into 'payments'. The Princess sincerely shared her journey.

A few points left a deep impression on me:
1. Infini wanted to enter the stablecoin space last summer; this kind of foresight and intuition is definitely at the T0 level of entrepreneurial teams.
2. If the business is related to Web2, it is inevitable to invest a lot of energy in compliance issues, and the costs required for compliance are not trivial. For example, some licenses are currently very difficult to obtain, and if you want to get them, you can only do so through company acquisitions.
3. Positive feedback can sometimes be a trap; one needs to be clear about their core objectives at all times. Infini initially hoped that launching the U Card would attract more users to increase the overall TVL, but they were blinded by the growth in user numbers and overlooked the fact that the TVL hadn't really increased, leading them to stray from this direction for a long time.
4. Do not pursue U Card entrepreneurship anymore. On one hand, the Princess's original words were that 'not a single penny was made'; on the other hand, the consensus we reached is that the U Card is a payment solution that everyone can only reluctantly accept in its current state, rather than a final solution. We look forward to the realization and popularization of stablecoin payments.
5. Infini has now returned to its original track and has completed its course correction; I am happy for them.

The complete interview content is in the comments section below 👇

Every carefully crafted product that ceases service is a pity. Especially in our industry, where we never prioritize product experience.

Since Day 1, I have been using Infini's products, learning and understanding the entire process of the U Card business along the way. Today, seeing Infini stop this business, I actually feel quite happy for them. It’s rare to see a team with such courage to pivot.

Boldly predicting, the U Card will eventually be replaced by formal banks, and true crypto payments will be funds sent directly from one’s own wallet in a self-custody model. Whoever can connect the offline payment channels will be the one to take a slice of this cake.

Not your keys, not your coins
嘎嘎帅
感谢老板 @Chloeppan @Foresight_News

Isn't it really sunny, and I heard it was taken in March or April?

It feels like the previous UNI airdrop, bringing back the taste of receiving airdrops initially.

The DeFi summer is about to return.

Since yesterday, when @Mira_Network announced it would allocate 0.5% of its tokens to Yappers, Twitter has been flooded with various project introductions. After reading a few articles, I still didn't understand what this project was about.

So I went to their official website and read their white paper from start to finish, and I want to share my understanding of this project with everyone.

Mira Network: Building a Trustworthy Verification Engine for AI

Have you ever thought that sometimes, when your AI is answering your questions, it actually has no idea whether what it's saying is true? It's just piecing together a sentence that seems plausible under the influence of certain weights.

In other words, it's seriously babbling.

This is not a problem of a specific model, but rather a fundamental ailment of the entire AI technology stack: hallucinations are unavoidable.

This means that before we dream of a future where AI makes autonomous decisions, writes code automatically, and treats diseases,

there's a Damocles' sword hanging over our heads:
Who will verify whether what AI says is true?

This is what Mira aims to do.

+ Break down all AI-generated content,
+ then have a decentralized verification network validate whether the results reach a consensus,
+ and issue a "truth determination certificate" with encrypted endorsement for each judgment output.

The whole process is like breaking an article down into sentences and having different people cross-check whether each sentence is correct.

What it relies on is not a specific authoritative model, but rather multi-perspective judgments from different AIs, honest participation driven by a reward mechanism, and the consensus results that ultimately form.

You can think of Mira as the "notary office" of the AI world, an intelligent verification machine that is always online, globally distributed, and combinable on demand.

Technical Logic: It’s not about who has a bigger model, but about letting multiple models correct each other.

We have all seen AI seriously babbling; it doesn't want to deceive, but the information in its head is already a mess.

Some of that information and data is outdated, some are contradictory, and some simply lack logic.

But Mira's solution is not to create a bigger model but to think from a different dimension:

Since no single model can achieve zero errors, let multiple models compete against each other to reach a consensus.

Incentive Mechanism: Verification is not voluntary work; honesty must be rewarding.

You might ask, if verification relies on models, why would nodes honestly participate and not engage in wrongdoing?

Mira has set up a relatively comprehensive mechanism to address this issue:

+ Verification tasks will be standardized into choices; if nodes guess randomly, they may not guess correctly;
+ Nodes must stake assets to participate in verification;
+ If answers frequently deviate from consensus, the system will trigger slashing, reducing the stake;
+ Guessing randomly is less profitable than doing honest work, which is the fundamental logic that allows the verification mechanism to endure.

As the network grows larger and model diversity increases, the window for cheating gets smaller.

Privacy Solution: Decompose + Shard, nodes cannot see the whole picture.

Mira is also very restrained in terms of privacy; it does not require nodes to see the user's original content. The system will first extract each "key statement" from the original content,

such as "Sleepy is a handsome guy" that clearly states right or wrong,

and then submit them for verification one by one, distributing them randomly. Any individual node cannot access the entire context, ensuring the minimum visibility of user data.

The subsequent roadmap will also introduce encrypted computation, such as MPC or ZK, gradually decentralizing content transformation logic, making the entire system increasingly transparent while exposing less data.

Ultimate Vision: Bringing verification into the core of AI.

Currently, all large models follow a "generate first, human review later" process. Mira aims to overturn this process, allowing future AIs to complete verification at the moment of generation.

When this model is established, AI output will no longer require human fallback, and artificial intelligence will finally evolve from a "generating tool" to a "trustworthy collaborator."

At that point, we can truly discuss AI autonomy and revolutionary change.

Not long ago, my kitten got sick. After taking it to the vet, I applied for pet insurance reimbursement. In addition to submitting various documents provided by the hospital, I also needed to take a photo of the kitten's nose to verify its 'nose print', which is quite interesting.

I asked and found out that the nose print is actually equivalent to the kitten's ID card, similar to human fingerprints, irises, palm prints, etc. To prevent the misuse of insurance limits, this verification mechanism was introduced.

Recently, I often come across the project @Humanityprot shared by others on Twitter, which authenticates whether you are a real person by verifying palm prints. Many people have published educational content and participation tutorials about this project, so I won't elaborate further; I just want to share my perspective.

In the era of AI, why do we need 'real-name systems'?

1｜From an anonymous utopia to an identity anxiety era

At the dawn of the Web, we embraced anonymity as freedom and a borderless world as an ideal. Forums, communities, cryptocurrencies, and even DAOs were all built on the foundation of 'decentralized identity'.

However, as we enter the AI era, this structure is rapidly becoming ineffective. We are facing a collective anxiety:

We can no longer distinguish whether the speaker is a human or a bot.

After 2023, social platforms like X, Reddit, and Discord frequently encounter bot abuse, with data farms operating on a massive scale; the Web3 world is also witnessing 'witch attacks', where one person splits into thousands of addresses to exploit airdrops and manipulate votes.

If you are a Builder, there are currently two fundamental problems that cannot be avoided:
+ You cannot know if the person you are talking to is a real person;
+ You cannot ensure that one person only participates once, rather than a thousand times.

This is not a Web3 issue; it is an identity crisis in the AI era.

Real-name systems, to some extent, are becoming necessary again.

2｜But traditional real-name systems are a misaligned solution

National ID cards, passports, mobile phone numbers, bank card bindings...

These are the real-name system infrastructure of the Web2 world.

But its biggest problem is that the binding of identity and rights is centralized, relying on governments, platforms, and banks. It addresses compliance issues, not authenticity issues.

More critically, it brings structural risks to privacy:
+ Centralized KYC data breaches are currently the largest source of data black markets on the internet;
+ Platforms abuse real-name systems for advertising profiling, credit control, and social sorting.

In the narrative space of AI + Web3, we need a new real-name system:
+ It only verifies that you are human without revealing who you actually are;
+ It does not expose information and does not store data;
+ It is not initiated by any country or platform but operates on a decentralized structure.

3｜ZK + Biometrics: not a simple KYC, but a 'decentralized real-name system'

Humanity Protocol and Worldcoin seem to be doing the same thing: using your 'body' to prove you are human.

It sounds scary, even evoking thoughts of totalitarianism, but the underlying mechanism is a complete reversal:

You are not handing over your privacy to the platform; instead, you are generating an irrefutable, untraceable, and tamper-proof 'evidence of existence'.

This relies on two key structures:
+ Biometrics: iris, fingerprints, facial features, are existential symbols that you cannot 'transfer';
+ Zero-knowledge proofs: only verify 'whether it is true', without disclosing 'specific content'.

This combination gives 'human identity' three important properties for the first time:
+ Scarcity (you can only have one self)
+ Verifiability (you can prove to any system that you are a real person)
+ Privacy (nobody knows who you are)

4｜Why is 'human proof' essentially a power structure?

Many people mistakenly think that Humanity Protocol is a 'anti-bot' tool.

In fact, it is a more fundamental structural declaration:
'Who is considered human' determines 'who has the right to participate'.

As AI sweeps through social platforms, content platforms, DAO governance, and even asset distribution, 'proving you are human' gradually becomes the threshold for entering the world.

This is not a security issue; it is a power issue.

It determines your eligibility to receive airdrops, participate in governance, sign contracts, gain reputation, and influence narratives.

It will shape a new social hierarchy on the chain:
+ Those without a human ID are the silent observers;
+ Only those with a unique ID can become 'digital citizens'.

In other words, we are moving towards an era of 'human assetization'; the fact that you are human itself becomes a resource that can be priced, proven, and empowered.

HYPE is already 40 dollars, are you still in cash?

Just discovered that today we finally broke 1000 Yaps

Recently, I've resumed serious content output, gradually increasing the account's weight, and I can't be slack anymore

So, is there any news about Kaito's next airdrop? 🥹

Let's talk about prediction markets and information finance, sparked by the collaboration between Polymarket and Twitter.

When information is assetized, when cognition can be bet on, and when opinions have a price, we are entering a market structure where gaming precedes truth, and cognition surpasses reality.

Prediction markets are no longer just speculative tools but have become the infrastructure for the financialization of information.

The following four keywords might help us reunderstand this change.

1｜Information Assetization

The essence of prediction markets is to price information.

In the traditional world, information is just 'content': you see it, you share it, you debate it, but you don't pay for it.

The emergence of prediction markets has made information for the first time a 'bettable event'; even if it hasn't happened yet or its truth is difficult to discern, as long as there is uncertainty, it can be packaged into bettable financial products.

Whether it's 'Will Trump win the election?', or 'Will a certain country go to war?', or even 'Will a certain celebrity get engaged?', all can have odds opened, liquidity gathered, and be priced by the market.

The boundaries of information are no longer determined by reality but by market interest.

2｜Financialization of Public Narratives

Arguing, voting, and sharing on X have always influenced narratives; and prediction markets simply turn that influence into price curves.

Every tweet, every social media spat, every viral video essentially influences the public's judgment of 'the likelihood of future events'.

Prediction markets transform this fluid cognitive competition into stable betting behavior:
Group sentiment = Direction of chips
Change in popularity = Fluctuation of probability prices
Manipulation of public opinion = A form of 'cognitive insider trading'

Information is no longer just information; it is a bettable asset.
Narratives are no longer just culture; they are a priceable liquidity pool.

3｜Reconstruction of Economic Incentives for User Behavior

Expression equals betting, information equals speculation.

On traditional platforms, expressing an opinion is free; the cost is just keyboard strikes and emotional output. But prediction markets change everything:
You have to bet on your opinion
You have to bear the risks of cognition
The way you express your stance is no longer commenting, but building a position

This mechanism brings about a new incentive for expression; you can make money by judging the direction of information, and you can also lose money for making incorrect judgments.

Users are no longer bystanders but active participants in the market, even price setters in the betting market.
A cognitive arbitrage era is quietly arriving.
4｜Tradability Over Truthfulness

Truth and falsehood are no longer important; whether someone believes it or whether it can cause a chain reaction is what matters.

This may be the most dangerous point.

In prediction markets, what is valuable is 'whether the event can attract a large number of bets', not 'whether the event actually happened'.

As long as someone places a bet, information can open; as long as emotions can be manipulated, rumors can become sources of volatility.

This is a highly financialized alchemy of attention:

Truth recedes, gaming rises
Objective facts yield to cognitive trends
What influences reality is not the event itself, but the market's pricing reaction to it

If you know what we are talking about, I recommend you research @buzzingdotclub
A next-generation prediction market combined with AI, and a winning project from the Virtuals Hackathon.
Stakeholders, clear signals, DYOR, I love Brother Luke @DeFiGuyLuke

LOUD, which once occupied over 70% of Kaito Mindshare, now has a market value of less than 3M, and there is hardly any related content on social media anymore.

Well, didn't the project team already say, 'This is a market experiment'? It's totally normal for an experiment to fail (Ma Di face).

To be honest, this whole project gives me the impression of being hasty, as if it was something thought up on a whim.

At the beginning, what attracted users to post content was the 'guaranteed profit expectation' based on token subscriptions, a pure arbitrage temptation. Many people posted content because they thought they could make money from it, and that’s it.

The premise for the project's economic model to function is to have KOLs continuously produce content and attract retail investors, but aside from the initial guaranteed profit opportunity, there’s nothing else for KOLs to post about. For the project team, this is a huge taboo in marketing; if you want others to promote you, at least you need to give them something to promote.

This is actually the fatal flaw of projects that we currently see using InfoFi for cold starts: when you make content production entirely reliant on incentives without building a sustainable content supply system, so-called popularity is merely 'consuming the future'. It looks prosperous on the surface but is actually just spinning its wheels; once expectations fade, all users will quickly withdraw because they never came here to stay from the beginning.

Nevertheless, I still believe that InfoFi is effective. It has been proven that even if a project is terrible, as long as the incentives are in place, the market crowd can still hype it up; if it were a legitimate project, it would definitely yield better results.

Its core value is the structural optimization of marketing: traditional marketing methods also require spending money, whether it's building your own marketing team, outsourcing marketing, or conducting marketing activities, money is spent regardless, so why not find a more cost-effective way?

The problem is not with InfoFi, but rather that the project itself was not prepared to handle the traffic brought by InfoFi. Users can be driven by incentives to produce content, but what happens after that? How is the story told? How does the experience keep up? How is volume converted into trust? These are the fundamental issues.

InfoFi can amplify the project's strengths but can also quickly expose its hollowness. If your product is terrible, no amount of incentives will help. Marketing is never a panacea; the true core competitiveness always lies in the quality and structure of the project itself.

A few days ago, I met friends from the Ton Foundation offline in Beijing, and during our chat, I updated my understanding of TON.

The biggest feeling is: the information gap is much more serious than imagined.

After the hype from the previous wave of mini-games faded, they are actually advancing very quickly globally, especially in the areas of DeFi and Sticker NFTs. However, there is almost no mention of this in the Chinese community.

Sticker NFTs have become one of the most active asset forms on TON, with exaggerated trading volume and user activity, yet no one here is talking about it. Meanwhile, there are still many Web2 game companies or studios in China researching how to launch mini-games on TG, connecting with TON through some Web2 gaming exhibitions. They may not understand the rise and fall of the previous wave of TG mini-games and are just trying a new growth path in the increasingly competitive Web2 environment.

There are indeed problems; TON is working quite hard, but TG finds it difficult to make significant updates or upgrades from a product perspective because the founder has a strong obsession with the product form, and this 'minimalist aesthetic fundamentalism' has become its ceiling.

Additionally, the development team is also quite small, disproportionately so, just as I had heard before. In such a structure, it resembles an ecosystem that grows on the margins with Telegram as its interface, and many narratives were not prepared for us from the very beginning.

In other words, the story of TON itself is not meant for us. This might still be an opportunity.

Today I want to discuss the recent matter of SUI recovering stolen funds.

Many people view it as a 'decentralized collapse'; at first, I thought so too, but after careful consideration, I believe it is not the case.

I think this is actually a complete demonstration of decentralized governance logic. It exposes the actual operating logic of the consensus mechanism and reminds us that decentralization is never the dictatorship of code, but rather the autonomy of consensus.

In the past, influenced by the founders of the early Crypto industry, we have been accustomed to 'Code is law', always treating 'technological neutrality' as a sense of security. But a decentralized network is essentially a process in which a group of nodes jointly maintains a ledger; its order does not come from the absolute authority of code but from the consensus of participants.

When a significant anomaly occurs at a certain moment: code errors, asset risks, systemic crises... those logics that were once regarded as 'immutable' will be discussed again, and the result of this discussion is called consensus.

Blockchain is not a dystopian paradise, but a continuation of the power structures of the real world. In this world, every node, every vote, every proposal can leave a record on the chain, no longer a black box.

Governance is politics; it is a game of humans. You cannot want compensation, emergency responses, and repair capabilities on one hand while expecting 'Code is law' on the other.

As mentioned earlier, I initially thought this was a decentralized collapse, but now I prefer to see it as a mirror.

It reflects the complex nature of decentralization and forces us to confront: the power of governance, the power of technology, and who we are willing to entrust our trust to. I will likely devote more energy to focus on and participate in the SUI ecosystem because this incident shows that SUI actually aligns more with my expectations and vision.

Actually, I don't really recommend you all to rush in after seeing a tweet about Opensea.

Given its nature, there is a high possibility that they won't release any airdrops by 2026. If you all rush in, they can just enjoy the profits, and they have even less motivation to issue tokens.

Although I also hope they release it soon, I should be able to grab a lot of airdrops, but reason tells me that it probably won't happen in the short term.