Authorization Before Execution: The Missing Layer in DeFi
I've started wondering whether DeFi has been solving the same trust problem in the least efficient way possible. Every new protocol spends time defining its own authorization rules, integrating separate security checks, and maintaining independent compliance logic before transactions can move. That approach may work for individual applications, but it doesn't naturally create reusable trust across an ecosystem. The more protocols that emerge, the more duplicated authorization infrastructure we end up maintaining instead of sharing. Newton Mainnet Beta made me look at authorization differently. Instead of leaving every protocol to build its own approval system, Newton introduces an onchain authorization layer where transactions are evaluated against active policies before settlement. The result isn't a public disclosure of sensitive data or internal policy logic. It's a cryptographically signed pass/fail attestation that proves the required policy was enforced. That distinction changes the role of blockchain infrastructure in a subtle way. For years, execution has been the network's primary responsibility. Once the required signatures exist, the transaction settles. Newton moves part of that responsibility to the authorization stage. Active policies for compliance, identity, security, and risk are evaluated before value moves, and the network records a signed pass or fail attestation instead of exposing the policy itself. I think that's a more scalable trust model because applications no longer need every participant to inspect complex authorization logic. They only need confidence that the required policy was enforced before execution. If authorization becomes standardized infrastructure instead of application specific logic, developers can spend less time rebuilding the same trust assumptions and more time creating better financial applications. The real advantage isn't reducing development effort. It's making trust itself easier to reuse. As more protocols rely on the same verifiable authorization framework, consistency becomes easier to achieve without requiring every application to expose sensitive policy details or recreate identical security models. What I find most interesting is what this could mean for the next generation of onchain applications. If developers eventually stop asking how to build authorization and start asking which policies to adopt, Newton won't simply introduce another infrastructure layer. It could change where trust is created in DeFi. Execution will always matter, but the protocols that define which actions are permitted before execution may end up shaping the ecosystem even more. #Newt @NewtonProtocol $NEWT
From Transaction Execution to Transaction Authorization: The Newton Protocol Approach
For years, DeFi innovation has been measured by one question: how efficiently can a transaction be executed? I'm beginning to think the next stage of infrastructure will be defined by a different question altogether. Should that transaction be authorized before it ever reaches execution? That shift is what makes Newton Mainnet Beta interesting to me. Traditional blockchain infrastructure excels at recording and executing transactions. Once the required signatures are present, the network processes them and stores the outcome onchain. But institutional capital, curated vaults, and complex DeFi operations often require something more. They need every transaction to satisfy predefined security, compliance, identity, and risk policies before assets move. Newton Protocol addresses this by introducing an onchain authorization layer. Instead of treating policy checks as an external operational process, it evaluates every transaction against active policies before settlement and produces a signed onchain pass or fail attestation. Authorization becomes part of the infrastructure rather than an afterthought. What stands out isn't just the technical design. It is the change in how trust is established. Instead of asking whether a transaction was successfully executed, participants can first verify whether it met the rules that were supposed to govern it. That creates a more transparent foundation for curated DeFi vaults and other policy driven applications. Whether this becomes the next standard depends on adoption. Developers must build around it, institutions must rely on it, and users must begin expecting authorization to exist before execution rather than after it. Execution made decentralized finance possible. Authorization may determine how confidently it continues to scale. @NewtonProtocol $NEWT #Newt
Beyond Billions: The Real Challenge for Curated DeFi Vaults
I used to think the biggest challenge for curated DeFi vaults was finding the best yield. As more institutional capital entered crypto, I assumed the winners would simply be the vaults with the smartest strategies. The more I looked, the more I felt that wasn't the real bottleneck. The harder problem isn't deciding where capital should go. It's proving that every decision respects the vault's risk policy before assets move. Today, curated vaults collectively manage billions of dollars, and that number continues to grow. Yet many of the controls protecting that capital still rely on fragmented offchain processes. Exposure limits, approved assets, allocation rules, compliance requirements, and emergency restrictions are often enforced through dashboards, internal reviews, multisig approvals, or operational procedures rather than onchain infrastructure. That creates an interesting contradiction. Blockchains are excellent at recording what happened. They're far less effective at proving a transaction should have happened in the first place. That's why Newton caught my attention. Instead of treating risk policies as documentation that humans promise to follow, Newton turns them into programmable rules that can be verified before execution. The transaction doesn't simply settle and get audited later. It first has to satisfy the policy itself. What makes this idea more interesting is that it isn't just theoretical. With the launch of VaultKit alongside Newton Mainnet Beta, the first deployment targets Euler vault curators on Ethereum and Base. Rather than asking vault managers to manually enforce every operational policy, VaultKit gives them a framework where authorization becomes part of execution itself. The surrounding ecosystem makes the picture even clearer. Newton's integration with Vaults.fyi allows live vault performance data to become an input for programmable policies. Instead of checking APY dashboards manually, developers can define conditions that must be satisfied before automated allocations occur. That transforms performance metrics from passive information into enforceable rules. The same philosophy extends beyond yield. Identity verification through Veriff can become a pre-transaction requirement rather than a separate onboarding checklist. Compliance signals, sanctions screening, and external data sources can all feed into authorization policies before funds move. Compliance shifts from paperwork to infrastructure. What I find most compelling isn't any individual integration. It's that each partner contributes a different piece of trust. Vaults.fyi supplies vault intelligence. Veriff supplies identity assurance. Together, they suggest Newton isn't trying to replace existing services. It's creating a common authorization layer where those services become cryptographically enforceable before execution instead of being verified afterward. Whether this becomes standard infrastructure will ultimately depend on adoption. Vault managers won't integrate another layer unless it improves security without adding unnecessary complexity. That's the signal I'll be watching. If more curated vaults begin embedding authorization directly into their execution flow, Newton's biggest contribution may not be another DeFi primitive. It may be proving that trust doesn't have to be assumed after settlement. It can be verified before settlement ever happens. #Newt $NEWT @NewtonProtocol $NFP $BTC
Most Protocols Tell You What Happened. Newton Decides What Can Happen Before It Does.
I used to think the next big improvement in DeFi would come from faster chains or lower fees. Lately, I've started questioning that assumption. The more protocols I explore, the more I feel we've become obsessed with making transactions faster while spending very little time asking whether every transaction should be allowed to happen in the first place. That question is what made me pay attention to Newton Protocol and its Mainnet Beta. What caught my attention wasn't another performance upgrade. It was the idea of treating authorization as infrastructure instead of an afterthought. Transactions integrated with Newton's authorization layer are evaluated against active policies before settlement. The result is a signed pass or fail attestation that can be verified onchain. To me, that's a different way of thinking about trust. Instead of explaining what happened after execution, the goal is to prove that the required policy was enforced before the transaction became final. I think that's an important distinction, especially as AI agents begin interacting with DeFi on behalf of users. The more decisions software makes automatically, the less comfortable I am relying on broad wallet permissions alone. Automation scales quickly, but so can mistakes. Clear authorization rules that are enforced before value moves feel like a more durable foundation than simply monitoring activity after it has already happened. That doesn't automatically mean Newton succeeds. Infrastructure only matters if builders decide it's worth integrating. If developers continue designing applications around this authorization model after Mainnet Beta, and users gain stronger security without noticing extra friction, that tells me the idea is solving a real problem instead of introducing another layer of complexity. That's the signal I'm watching. Not transaction volume, but adoption. Will developers start treating authorization as a core building block instead of a security feature? Will other protocols begin relying on verifiable policy attestations? If that happens, Newton's biggest contribution may not be faster execution at all. It may be changing when trust is established, before a transaction settles instead of after it has already become part of the blockchain. $NEWT #Newt @NewtonProtocol
What Is Newton Protocol? Founders, NEWT Circulating Supply, and Why It Caught My Attention
I almost ignored Newton Protocol. Not because it looked bad, but because I've seen too many crypto projects promise to redefine Web3. Faster chains. Smarter AI. Better user experience. After a while, those words start sounding the same. So instead of reading posts about Newton, I opened the documentation. The first thing I realized was that Newton isn't really competing to build a better blockchain. It's asking a different question. What happens when AI agents become normal users of blockchain? At first, I was focused on how AI could become more capable. But after spending time with the documentation, I realized capability wasn't the part that concerned me the most. The real problem is authorization.
I caught myself thinking about how much time I spend doing the same things onchain. Swapping tokens here, claiming rewards there, signing another transaction, moving assets around. It's not hard work, just repetitive. The more I thought about it, the more I realized that having AI handle these routine tasks doesn't actually sound that strange anymore. How do you give software enough freedom to be useful without giving it enough freedom to become dangerous? Newton's answer isn't asking users to trust AI more. It's reducing how much trust is required in the first place. Instead of unlimited wallet permissions, users create programmable authorization policies that define exactly what an AI agent is allowed to do. Every request is checked against those policies before execution. If an action falls outside the rules, it simply doesn't happen. That idea felt surprisingly simple, yet I hadn't been thinking about it before. I also found Newton's architecture more interesting than I expected. Authorization is separated from execution, so policy logic isn't tied to a single blockchain. As more applications become multichain, reusing the same authorization framework across different networks feels much more practical than rebuilding permission systems over and over again. Another detail that made me spend more time researching was the team behind the project. Newton originated from the team behind Magic Labs, co founded by Sean Li and Jaemin Jin. Their background in embedded wallet infrastructure doesn't guarantee success, but it explains why user authorization seems to sit at the center of Newton's design instead of feeling like an afterthought. Then I looked at NEWT. Price wasn't the first thing that interested me. Circulating supply, allocation, future unlocks, and incentive design usually tell me much more than a chart. I've made the mistake before of paying attention only to price. Now I spend more time asking whether the economic model can survive once the excitement fades. One technical decision I genuinely liked was Newton's use of Rego for defining authorization policies. Most users won't ever write Rego themselves, and honestly they shouldn't have to. What matters is the shift in thinking. Wallet approvals become programmable rules instead of permanent permissions. That feels much closer to how AI should interact with financial systems. The biggest thing Newton changed for me wasn't my opinion about AI. It changed my opinion about infrastructure. I've spent a long time assuming the next race in Web3 would be about faster execution, cheaper transactions, or more powerful AI models. Now I'm not so sure. If AI agents become normal participants in blockchain networks, authorization may quietly become one of the most important layers nobody talks about today. I still don't know if Newton Protocol will become that layer. Adoption, developers, and real usage will decide that, not ideas alone. But it's one of the few projects that left me thinking long after I finished reading the documentation. And in crypto, that's usually a much stronger signal than hype. $NEWT @NewtonProtocol #Newt $TAC $SOL #bnb #Aİ
The deeper I look into OpenGradient, the less interested I become in comparing AI models or blockchain performance. That's how I started evaluating the project, but it no longer feels like the right lens. Better models will be replaced. Faster infrastructure will eventually become standard. The harder problem is building an ecosystem where developers keep building, users keep returning, and AI applications continue creating demand after incentives disappear.
That's why I don't see Season 2 as the destination. I see it as the onboarding layer. Almost every crypto project can attract users through rewards, but campaigns eventually end. The real question is what happens next. Do developers continue launching applications because the infrastructure is genuinely useful? Do users keep spending credits because AI services solve real problems? Retention says far more about a network than participation ever will.
The architecture also makes more sense through that lens. Cosmos SDK provides AI native flexibility, while EVM compatibility lowers the barrier for Ethereum developers. Instead of choosing between specialization and accessibility, OpenGradient appears to be trying to combine both. That may matter more than another performance upgrade.
I also stopped seeing x402, MemSync, Model Hub, PIPE, Twin.fun, and AlphaSense as isolated products. Together they look like a shared trust architecture. Vanilla inference optimizes for speed, while TEE backed execution and ZKML reduce how much trust users and developers must place in a single system. Those don't feel like only technical improvements. They also change the economics of adoption.
OpenGradient uses Seedream 5.0 Lite and 4.5 to turn text prompts into AI-generated images. Users can create artwork, illustrations, concepts, and designs while keeping prompts and images private.
I'm watching behavior, not benchmarks. If builders and users stay after Season 2, genuine network effects could drive the future $OPG economy.
#opg $OPG @OpenGradient I opened OpenGradient Chat expecting another AI product. I came away thinking more about trust than model quality.
For years, using AI has quietly meant accepting a tradeoff. The more personal the question, the more trust you place in the platform handling it. Medical concerns, legal questions, financial decisions, and private thoughts often become linked to an account and stored somewhere beyond the user's control.
OpenGradient Chat challenges that assumption. Rather than asking users to trust a company, it embeds privacy into the system. Messages are encrypted before leaving the device, Oblivious HTTP separates identity from content, and prompts are processed only inside Trusted Execution Environments (TEEs) with remote attestation. The goal isn't simply to claim privacy, but to make key privacy guarantees independently verifiable.
What I find more interesting is what this could mean for AI infrastructure. OpenGradient Chat applies the same privacy-preserving and verifiable AI architecture that underpins the OpenGradient network. Specialized GPU nodes perform AI computation, while dedicated TEE nodes enable confidential execution with remote attestation, making key execution and privacy guarantees independently verifiable. The result is a consumer application built on the same principles as the network itself rather than a standalone chatbot.
The product still delivers a familiar AI experience through access to frontier models like ChatGPT, Claude, Gemini, Grok, and ByteDance Seed, alongside web search, file uploads, image generation, and a roadmap that expands those same privacy guarantees to image and video models.
The signal I'm watching isn't how many models OpenGradient adds next. It's whether users begin choosing AI platforms based on verifiable privacy instead of benchmark scores alone. If that behavior changes, trust may become infrastructure rather than marketing-and that could matter just as much as the intelligence itself.
I used to think OpenGradient Chat was simply another AI assistant. After reading the whitepaper and exploring the platform, I realized the bigger opportunity isn't just the models-it's the infrastructure connecting AI, payments, and developers.
Through OpenGradient's TEE-verified infrastructure, users can access AI models from providers including OpenAI, Anthropic, Google, and xAI, while the OpenGradient Model Hub enables deployment and inference of open-source models on the network. Private Chat also offers the Nous Hermes model for users who prefer a more open conversation experience. The platform supports AI image generation through supported image-capable models, allowing users to create images directly from text prompts without switching between different tools.
What changed my perspective is the architecture underneath. x402 is designed to enable payment-gated AI inference, PIPE enables on-chain machine learning execution, and OpenGradient's products aim to give developers a unified environment instead of stitching together separate AI, payment, and infrastructure services.
The project is backed by investors and ecosystem partners including a16z crypto, Coinbase Ventures, SV Angel, Foresight Ventures, Symbolic Capital, NEAR, and Celestia. While backing doesn't guarantee success, it does show the project has attracted support from established names across AI and Web3.
I also find the incentive model interesting. Purchasing OpenGradient Chat credits and actively using the platform is one of the activities recognized in the Season 2 OPG airdrop campaign. That aligns rewards more closely with real platform participation instead of focusing only on passive token ownership.
The question I'm watching is whether developers begin treating OpenGradient as the infrastructure layer behind their AI applications rather than just another AI chat platform. If that happens, the network's economics could increasingly be driven by real AI usage instead of speculation alone.
I used to think blockchain architecture was mostly about speed and scalability. The more I looked at OpenGradient, the more I realized the bigger challenge might be balancing specialization with accessibility.
Most chains choose one extreme. They either build highly customized infrastructure that offers unique capabilities but creates adoption friction, or they stay close to Ethereum standards and inherit its limitations.
What makes OpenGradient interesting is its attempt to combine Cosmos SDK flexibility with EVM compatibility. That creates room for AI-native features while still allowing developers to use familiar Ethereum tools.
After spending time with OpenGradient Chat, I started viewing it as more than a chatbot. Each interaction is a small test of whether decentralized AI can generate real demand instead of depending purely on market narratives.
The same thought applies to the S2 airdrop. Bringing users into an ecosystem is relatively easy. The harder question is how many remain active once incentives disappear. Retention often says more about product value than participation numbers.
That also connects to OPG economics. The most important metric may not be how many people hold the token, but how many AI interactions, services, and applications eventually depend on it. If usage grows, utility and demand become linked in a much stronger way.
For me, the real experiment isn't whether OpenGradient can build AI-native infrastructure. It's whether it can keep adding advanced AI functionality without losing the accessibility that attracted developers in the first place.
If decentralized AI becomes more specialized over time, can OpenGradient maintain that balance between flexibility, usability, and sustainable demand?
I went into OpenGradient Chat expecting to compare AI models.
Instead, I left thinking about trust.
The obvious story is the model access. OpenGradient was among the first platforms to integrate Claude Fable 5, a model designed for longer conversations and stronger contextual understanding. At the same time,Users can generate images too with Image Studio live with OpenGradient Chat ,while Private Chat includes Nous Hermes. That feels less like a model strategy and more like a retention strategy. Users rarely stay because of one model. They stay because a platform solves multiple needs without forcing them to leave.
But that wasn't what kept my attention.
What stood out was the decision to treat privacy as infrastructure rather than a promise. Messages are encrypted on device, identities are stripped before requests reach a model, and trust is enforced through cryptography and hardware instead of a privacy policy.
That's why the S2 OPG airdrop caught my attention. Eligibility comes from purchasing credits and actually using them on the platform. The real test isn't how many users arrive because of rewards. It's how many remain after those rewards disappear.
If users keep returning, OpenGradient's biggest advantage may not be its models.
Maybe I'm looking at OpenGradient differently than most people.
Everyone is focused on the S2 airdrop.
I'm trying to understand the economy that could exist after it.
An airdrop is simple.
You complete tasks, earn rewards, and eventually the campaign ends.
An economy is different.
An economy is what happens when users, developers, AI applications, and network activity all start creating value for each other.
That's why the idea of a future $OPG economy is interesting to me.
Imagine thousands of users spending credits to access AI services.
Developers building new AI applications.
More applications attracting more users.
More users creating more demand across the ecosystem.
That's a real economic loop.
The stronger the network becomes, the more valuable participation can become for everyone involved.
S2 might be the incentive that brings people into OpenGradient.
But the economy is what could make them stay.
That's why I'm spending less time thinking about the size of the airdrop and more time thinking about the size of the ecosystem that could emerge around it.
Because in the long run, ecosystems create more value than campaigns ever can.
I didn’t think much about trust in AI systems until I noticed a simple but unsettling pattern. The same prompt can produce the same looking answer yet the way that answer is generated can be completely different underneath. That hidden layer is where the real question sits.
Most AI today runs on what you could call vanilla ML. It is fast, cheap and everywhere. You send input, you get output and you assume the system behaved correctly. But there is no proof of execution, no visibility into what actually happened between input and result. Trust is implicit, not earned.
Then comes Trusted Execution Environments TEEs. Here the model runs inside secure hardware isolated from external interference. It feels stronger because the environment is locked down. But the nature of trust does not really change, it just moves from software to silicon. You still do not observe the computation, you only trust the enclosure.
Zero Knowledge Machine Learning ZKML shifts the direction entirely. Instead of asking you to trust the output or the environment, it attaches a cryptographic proof that the computation was done correctly. You do not see the process but you can mathematically verify it happened as claimed. The tradeoff is cost, latency and complexity but the idea of don’t trust, verify becomes real in AI.
This is where the spectrum matters. Not every system needs full cryptographic assurance and not every system can afford it. Vanilla ML optimizes speed. TEEs optimize secure execution. ZKML optimizes verifiability.
The real shift is not choosing one winner but accepting that trust is no longer binary. It is layered. And the intelligence of future systems will depend not just on what they answer but on how provably that answer can be trusted.
The more I learn about AI, the less I believe intelligence is the real challenge.
Trust is.
Today, AI can analyze markets, generate research, write code, and influence decisions in seconds. What it usually can't do is prove how it arrived at those decisions. We get the answer, but the reasoning often disappears inside a black box.
For simple tasks, that might not matter.
For systems handling money, healthcare, infrastructure, or governance, it matters a lot.
Imagine an AI agent recommending a major portfolio adjustment during a market downturn. The recommendation could be brilliant. It could also be completely wrong. Either way, most users have no practical way to verify how that conclusion was reached. They are left with a choice between blind trust and complete skepticism.
Neither is a great foundation for the future of AI.
That's one reason OpenGradient stands out to me.
What caught my attention wasn't the promise of building smarter models. It was the idea that AI outputs should be verifiable rather than simply accepted. In OpenGradient's vision, trust is not treated as a marketing slogan layered on top of AI. It becomes part of the infrastructure itself.
The concept is surprisingly simple. Instead of asking users to trust a result because a system produced it, create mechanisms that allow those results to be independently verified. The difference sounds subtle, but it changes the entire relationship between humans and AI.
I also think this becomes more important as AI agents become increasingly autonomous. Intelligence alone won't be enough. Systems that manage value, information, and decision-making will need accountability built into their foundations.
The more I think about it, the more I feel the next AI race won't be won by the model with the highest benchmark score.
It will be won by the systems people trust when real value is on the line.
Because intelligence can capture attention.
But verification is what turns technology into infrastructure.
And infrastructure is what people ultimately depend on. $OPG #opg
Most AI projects are trying to make machines smarter. OpenGradient is making a different bet.
It assumes the future problem may not be intelligence. It may be verification.
The more I thought about that idea, the more interesting the project became. Today, when we use AI, we usually trust the result without questioning what happened behind the scenes. We trust that the model ran correctly, that the output wasn't changed, and that everything worked as expected.
But what happens when AI starts handling more important tasks?
At that point, trust alone may not be enough.
This is the problem OpenGradient is trying to solve. Instead of focusing only on AI capabilities, it is building infrastructure that allows AI outputs to be verified. In simple terms, the goal is to provide proof, not just answers.
That's where the OPG token fits into the picture.
What I like about the design is that OPG isn't presented as an extra feature attached to the network. It sits at the center of the ecosystem. As developers use AI services, verification tools, and other network resources, the token helps coordinate activity between participants and supports the system running underneath.
It also plays a role in staking and governance, helping secure the network while giving the community a voice in its future development.
But the most interesting part isn't the token itself.
It's the idea behind it.
Blockchain was built to verify transactions. OpenGradient is applying a similar mindset to AI. Instead of asking people to trust a model, it wants trust to be backed by proof.
If AI becomes a major part of the digital economy, that shift could matter more than many people realize.
Because in the long run, the most valuable AI product may not be intelligence.
It may be the ability to prove that the intelligence can be trusted.
I keep coming back to one simple thought in crypto.
Most capital doesn’t fail because it enters the wrong place. It fails because it doesn’t stay useful after it enters.
For a long time, I treated liquidity the same way most people do. You find a protocol, deposit, earn rewards, and then rotate to the next opportunity. It feels normal because the entire market is designed around short-term incentives.
But the more I observe different ecosystems, the more I realize something important is missing from that cycle.
Sustained usefulness of capital.
That’s where projects like Bedrock made me pause and think differently.
Instead of focusing only on attracting liquidity, the idea seems to lean toward keeping that liquidity active inside the system. Not just parked. Not just farmed. But continuously contributing as the ecosystem evolves.
And honestly, that small shift changes the mindset completely.
Because when liquidity is always working, participation stops feeling like a one-time decision. It starts feeling like being part of an ongoing system where value doesn’t reset every time incentives change.
We’ve all seen the opposite version too many times.
High rewards bring capital in quickly, narratives build up, and everything looks strong on the surface. But the moment incentives slow down, liquidity leaves just as fast. The cycle resets again.
That’s why sustainability matters more than hype.
For me, the real question isn’t “which protocol pays the most today.”
It’s “which system can keep capital engaged even when attention fades.”
Because in the long run, the strongest ecosystems won’t be the ones that attract liquidity the fastest.
They’ll be the ones that give liquidity a reason to stay active without being constantly pulled by the next incentive.
And that changes everything about how we think about participation. #bedrock $BR @Bedrock
Lately, I've been thinking about a different question.
Why is so much Bitcoin still sitting idle?
Bitcoin has become one of the most valuable assets in the world. People hold it, institutions accumulate it, and long-term investors treat it like digital gold.
But when you look deeper, most Bitcoin isn't actually doing anything.
It's stored.
Protected.
Waiting.
And that made me wonder if the next big opportunity in crypto isn't creating more Bitcoin but making existing Bitcoin more productive.
That's one reason Bedrock 2.0 caught my attention.
What interests me isn't simply the idea of earning yield. We've seen plenty of projects focus on that.
What stands out is the broader vision of helping Bitcoin capital work across multiple opportunities instead of remaining inactive.
The model combines different approaches, including quantitative strategies, lending markets, BTCFi participation, and exposure to real-world asset opportunities.
The way I understand it, each part of the ecosystem serves a purpose.
uniBTC acts as the capital layer.
Vaults create opportunities for deployment.
BRClaw adds an intelligence layer.
And $BR helps connect users to the ecosystem and its benefits.
What I find most interesting is the shift in mindset.
For years, Bitcoin's role was simple: buy, hold, and wait.
Now the conversation is starting to evolve.
People are exploring how Bitcoin can remain a strong store of value while also becoming more capital-efficient.
Whether this becomes the future of BTCFi or not, I think it's a trend worth watching.
Because the next chapter of Bitcoin may not be defined only by price appreciation.
It may be defined by what Bitcoin can do while it's being held.