Shehryar Shah g

Honnêtement, j’en ai marre de voir un projet de blockchain après l’autre prétendre résoudre le problème de « l’IA ». D’abord les applis DeFi, puis les solutions de couche 2, et maintenant le trading automatisé piloté par l’IA. Le dernier en date, c’est Newton Protocol (NEWT), présenté comme un « rollup sécurisé pour des stratégies pilotées par l’IA, des systèmes de trading automatisés et un marché pour les développeurs d’IA ». D’accord… exactement ce dont on avait besoin : encore un assemblage de buzzwords crypto. Après des années de cycles crypto, de promesses et de déceptions, je ne peux pas m’empêcher de lever les yeux au ciel un peu. Chaque vague apporte son lot de mille nouveaux tokens et de grandes promesses : plus rapide, plus intelligent, le tout présenté comme étant soutenu par l’IA dans le pitch. Pourtant, une fois la poussière retombée, la plupart d’entre eux s’éteignent discrètement. Soyons honnêtes : on a déjà parcouru cette route. Rappelez-vous tous ces « tueurs d’Ethereum » ? Ou la parade sans fin de bots on-chain qui jurent de nous rendre riches ?

Honestly, after spending years watching crypto go through the same emotional roller coaster over and over again, it gets harder to believe every new story. One cycle it's DeFi, then NFTs, then metaverse projects, then meme coins, and now apparently everything has AI attached to it. Every few months there's another token promising to change everything, another influencer calling something the future, and another community acting like history suddenly stopped existing. Let's be real, we've seen this movie enough times to know that most of the excitement disappears long before the product actually proves itself.
That's probably why Newton Protocol caught my attention in a strange way. Not because it looked flashy, but because it didn't. It is trying to build infrastructure around AI-driven automation and on-chain strategies, and infrastructure has never been the glamorous side of crypto. It's usually the boring stuff that people ignore until they suddenly realize they needed it all along. Whether Newton ends up becoming one of those projects is another question entirely.
The problem they're trying to address is real, though. If AI is going to handle financial decisions, execute trades, or interact with blockchains without constant human approval, there has to be some way of making those actions predictable and accountable. Blind trust has never worked well in crypto. We've already watched too many protocols fail because people assumed code, automation, or incentives would magically solve human problems. They usually don't.
The AI part is where I become cautious. Every company today seems to believe adding two letters to their pitch instantly makes them more valuable. AI is powerful, but it's also unpredictable. That's the part that worries me. Giving software more responsibility sounds exciting until it makes a decision nobody expected. Maybe the safeguards work. Maybe they don't. Until systems face real users with real money on the line, nobody truly knows how reliable they are.
Then there's the token, NEWT. I always ask the same question whenever another project launches one. Does it genuinely need a token, or is the token simply expected because that's how crypto fundraising has worked for years? Sometimes tokens have an obvious purpose. Other times they feel like an extra layer wrapped around a product that might function perfectly well without one. I don't know where NEWT will eventually fall on that spectrum, and I think it's healthy to admit that instead of pretending every token has unquestionable value.
Adoption is another thing people often gloss over. Building technology is difficult, but convincing developers, businesses, and users to change their habits is even harder. Crypto has produced plenty of technically impressive projects that barely anyone uses. That's a reality this industry rarely likes talking about because adoption doesn't happen through hype alone.
Maybe that's why I find Newton Protocol interesting without feeling excited about it. It isn't selling a fantasy about replacing the financial system overnight. It's trying to solve a practical problem that could become increasingly important if AI continues becoming part of crypto infrastructure. That doesn't guarantee success, and honestly, nothing in this industry does.
I've been around long enough to know that good ideas can fail just as easily as bad ones somehow become billion-dollar narratives. Maybe Newton Protocol works, maybe it doesn't. For now, it simply feels like one of those projects worth watching from a distance rather than celebrating too early, and after everything crypto has taught me over the years, that feels like the most reasonable position to take.
@NewtonProtocol #Newt $NEWT

Je suis dans la crypto depuis assez longtemps pour savoir que chaque cycle apporte une nouvelle histoire que tout le monde finit par croire. Il y a quelques années, c’était la DeFi, puis les NFT, puis les projets de métaverse, puis les actifs du monde réel, et maintenant c’est l’IA. Chaque semaine, un autre token prétend construire le futur, et chaque influenceur a, d’une manière ou d’une autre, une certitude totale sur le fait que celui-ci est différent. Honnêtement, à force, c’est épuisant. Le marketing change, les logos deviennent plus propres, les buzzwords deviennent plus compliqués, mais le schéma change rarement. La plupart des projets promettent bien plus qu’ils ne livrent, et les gens qui crient le plus fort disparaissent généralement quand l’enthousiasme retombe.

Honestly, I’m tired. Every few months a new buzzword coin pops up claiming to solve all our problems. First it was DeFi, then NFTs, now **AI-driven DeFi** – same story, new packaging. Newton Protocol ($NEWT ) is the latest marquee. It’s pitched as a “secure rollup” for AI trading bots and on-chain policy enforcement, an “authorization layer” for every crypto transaction. The marketing is slick: TEEs, ZK proofs, compliance, institutional guardrails – even tie-ins to millions of wallets via Magic Labs. But let’s be real, after *so many* cycles of “next big thing,” my cynical antennae are raised.
Newton’s literature is full of buzz. For example, coinmarketcap’s AI guide calls it a “verifiable, programmable compliance” layer that bridges smart contracts with off-chain realities. A CryptoTotem summary gushed that Newton “turns automation from a point of failure into a foundation of trust” by combining hardware enclaves and ZK-proofs. Magic Labs – the wallet startup behind it – even bragged they’re plugging Newton into 200K developers and 50M wallets to bake “compliance-as-code” into transactions. On Twitter they talk about building “the locks” for crypto’s glasshouse. It all sounds cutting-edge, until you realize it’s mostly corporate buzz.
At this point, I have to ask: **Does the world *need* this?** Sure, autonomous bots and AI trading are becoming a thing. And yeah, handing keys to a dumb bot has obvious risks: it could get hacked or go rogue. Newton is basically selling a permission slip for your robot, insisting the agent can only act within the cryptographic box you draw. One commentator put it well: Newton isn’t trying to make AI smarter – it’s trying to *contain* it. Instead of “trust the agent and verify later,” their pitch is “the agent physically *can’t* act outside the box you drew”. That *is* a novel framing. In theory, it means you get an attestation on-chain that “yes, your trade obeyed the rules” – a nice audit trail of compliance before any transaction clears.
But here’s the rub: such a solution only matters if you set the rules **right**. Every expert I’ve seen knock on Newton agrees: fancy tech won’t stop a bot from doing something dumb if you’ve given it broad permissions. As one skeptic noted, “Permission boundaries are only as good as how they’re defined. If a dev sets sloppy or overly broad permissions (because it's easier or the UI nudges them that way), you’ve basically recreated the same risk… just with extra crypto steps in between.” In plain English: “verifiable automation” doesn’t automatically mean “safe automation” – it just means you *can* prove what happened within whatever rules you set. Garbage-in, garbage-out. And in practice, most devs just want to get something working; will they really dial in perfect guardrails, or just allow-everything for speed? Good infrastructure that folks configure lazily isn’t much better than bad infrastructure.
Now, on to *trust*. Newton *talks* a big game about trustlessness and public auditability, but the reality is murkier. Recall that one of their big selling points is “zero trust” – no more blind reliance on oracle oracles or devs. Yet guess what? As of mid-2025 the core Newton code (the “Keystore” rollup and model registry) wasn’t even public. The team basically said “don’t worry, we’ll release the code after it’s ready.” Uh, right – trust us until then. That’s a big red flag in crypto: the very layer claiming to enforce trustless policies has itself been cloaked. A crypto blogger wrote about this paradox: *“the gap between a protocol promising trustless execution and a codebase that cannot yet be independently reviewed is not a small one”*. In other words, if they insist on giving guarantees only through code, why hide it? (They do have audits of the token contracts, but curiously *not* the main system.) It’s like selling a metal detector that finds bombs – but you’re not allowed to see inside the detector.
And what about those magic buzz numbers? Magic Labs likes to shout that Newton will bring compliance features to “200K developers and 50M wallets” overnight. Maybe – but marketing headcounts aside, I see no sign most of those wallets are using it. To date, the chatter around NEWT is *mostly* about its lofty vision, not real traction. CoinMarketCap’s AI bot analysis even notes that social volume is “quiet on price but loud on vision,” and trading activity looks more like unlock speculation than real usage. One user checked the DEX volume: a few million dollars in 24h, and wonders if that’s just anxiety around token unlocks (given another big unlock is coming) rather than fees flowing from actual policy checks. In short, the token is trading like a “security-token wearing fee-token language” – basically speculation on promises.
Let’s talk tokenomics for a second. NEWT has a 1 billion max supply, but only about **21.5% of that is unlocked** right now. The rest is cliff-vested over years. In fact, CoinMarketCap noted 139 million NEWT (over a third of circulating supply) was scheduled to unlock on June 24, 2026. That’s a classic short-term bearish catalyst: suddenly a lot of tokens hit the market. (Tokenomist confirms a ton is still locked up until 2029.) I’m always wary when a new “infrastructure token” hits exchanges with huge scheduled dumps. For a project relying on deflationary tokenomics or network fees, these unlocked dumps can swamp the exchange price unless genuine demand soars. Right now, demand seems to be… a handful of speculators and a lot of hype-speak.
Another nit: they say policies are executed by a decentralized operator network with cryptographic proofs. Great, but the *data* these policies use – sanction lists, risk scores, identity checks – still comes from somewhere. Newton’s blog admitted they’ve onboarded a few providers (Chainalysis for sanctions, Credora for risk, etc.). The user-facing sell is “compliance-as-code,” but underneath it’s *five companies instead of one compliance officer*. One comment nailed this: the on-chain proof only confirms “the policy ran correctly against whatever those providers said” – it never proves the providers’ data was right. In other words, you’ve just moved the trust to a few oracles. Crypto purists will note: you still trust Chainalysis on what “sanctioned” means, or Massive/Polygon.io on treasury curves. The blockchain magic only ensures they didn’t compute wrong given their inputs. “Verifiable” does not mean “objective truth,” it just means you can all check the math.
Beyond tech and tokens, there’s the **adoption problem**. This is niche stuff that only big institutions really care about (and regulatory whales). Crypto retail won’t rush to wrap their Uniswap bots in a Newton policy. If anything, they will yawn at more moving parts. The real users Newton needs are institutional DeFi builders – but guess what, those guys already often use centralized oracles and compliance on traditional rails. Newton claims to target “lockchain-facing institutions,” but convincing a bank to switch off-chain compliance into on-chain smart contracts is an uphill battle. Yes, Magic Labs’ integration blurb says devs can now apply real-time sanctions checks or KYC checks in transactions, which sounds neat. But who’s actually coding those checks into every NFT sale or DEX swap in practice?
At the end of the day, my brain says *maybe this works in a specific corner of crypto (like institutional RWA oracles)*, but *everything about the launch screams caution*. The team behind Newton (Magic Labs) isn’t anonymous – they have real users from their wallet business – so it’s not a random rug. But they’re essentially telling devs: “Go ahead, trust our system before seeing the code.” That’s a hard ask. In crypto you *demand* your code be open. All I see right now is grand visions and clever demos (like plugging U.S. Treasury curves into pre-trade checks), which is interesting but doesn’t guarantee someone will actually use it.
I’m not saying Newton is a scam. The problem it tries to solve – “what if an AI misbehaves with your money?” – is real. On-chain automation without guardrails is dumb. But remember the last time someone promised blockchain to solve X? Too many times we’ve heard “this will change the world” only to see an empty site. The phrase “onchain agent marketplace” or “AI agent” already makes me roll my eyes – it’s the same tricky phrase we’ve heard around dozens of memecoins and SDKs.
So, I’ll watch from a distance. I’ll believe it more when I see their *Explorer* showing tons of real policy checks, not just LinkedIn posts. Right now, it feels more like another cycle’s worth of hype wrapped in fancy tech lingo. Maybe Newton will become the Google of blockchain compliance someday, or maybe it’ll be just another cautionary tale of “be careful what you trust.” Until then, let’s be real: lots of shiny infrastructure projects sound nice on paper, but the crypto market has a way of punishing the unproven.
@NewtonProtocol #Newt $NEWT

Honestly, I don't know how many more crypto narratives I'm supposed to get excited about. After being around for a few market cycles, everything starts looking familiar. Every year there's a new trend that suddenly becomes the answer to every problem. First it was DeFi. Then NFTs. Then metaverse projects that promised digital worlds most people never visited twice. Now it's AI. Apparently every project has AI now, every founder is talking about autonomous agents, and every influencer somehow knew it was the future before everyone else. I've watched this movie enough times that it's hard to react with anything other than a tired shrug.
That's probably the biggest problem for projects launching today. Even if someone is building something genuinely useful, they're entering a market where trust has already been burned too many times. Crypto isn't short on ideas. It's short on patience. People have been promised financial revolutions, internet revolutions, gaming revolutions, and social revolutions. Some ideas worked. Most didn't. The hype always arrived much earlier than the actual product.
That's why Newton Protocol caught my attention, although probably not for the reasons its supporters would like. I wasn't looking for another AI token. I was actually wondering whether anyone was trying to solve one of the quieter problems that has been sitting in crypto for years. Automation sounds great until it controls your wallet. AI sounds impressive until it makes decisions involving your money. Then suddenly trust becomes a much bigger issue than intelligence.
From what I understand, Newton Protocol is trying to build infrastructure that allows AI-driven strategies and automated actions to happen under predefined rules instead of giving unlimited authority to software. On paper, that sounds reasonable. Maybe even necessary. If crypto keeps moving toward automation, then someone eventually has to figure out how people stay in control without manually approving every single action. That isn't the most exciting problem in crypto, but sometimes the boring problems are the ones worth solving.
Still, let's be real. Good ideas don't automatically become successful projects. Crypto history is full of technically smart teams that disappeared because nobody actually used what they built. That's the part people forget. Building infrastructure is rarely glamorous. Nobody tweets about secure authorization systems with the same excitement they reserve for meme coins doing ten times overnight. Infrastructure usually matters only after something breaks.
The AI part is where I become much more cautious. Maybe I'm just getting older in crypto years, but I've stopped believing that adding AI automatically makes something better. AI can process information quickly, but that doesn't mean I want it making financial decisions without extremely clear boundaries. Markets are unpredictable enough when humans are involved. Adding AI doesn't magically remove risk. In some cases it simply creates new kinds of risk that nobody fully understands yet.
That's the part that worries me. Everyone keeps talking about autonomous agents as if trust is already solved. It isn't. If an AI strategy loses money because markets move unexpectedly, that's unfortunate. If it loses money because permissions weren't designed properly or because someone exploited the system, that's a completely different conversation. Security usually doesn't become exciting until after the first disaster.
Then there's the token itself. Every crypto project eventually introduces a token, and every token somehow manages to have governance, utility, staking, incentives, and ecosystem participation all wrapped into one package. I've heard those explanations more times than I can count. The real question isn't whether a token has listed utilities. The real question is whether people would still want to use the protocol if the token didn't exist. That's a much harder question, and one that often doesn't get answered until years later.
I'm also curious about adoption because that's where reality usually interrupts the story. Developers might appreciate better infrastructure. Traders might like safer automation. AI builders might enjoy having a dedicated marketplace. But wanting something in theory and using it every day are completely different things. Crypto users are incredibly demanding. If something feels complicated, they'll ignore it. If something feels risky, they'll avoid it. If something doesn't immediately offer an obvious benefit, they'll move on to the next shiny trend.
Maybe that's unfair, but that's the market we've built. Attention moves faster than products. Narratives spread faster than adoption. Sometimes it feels like people invest more in headlines than in technology. That's why I don't immediately celebrate projects simply because they're connected to AI. We've already seen enough buzzwords come and go to know that marketing can only carry a project so far.
What I do appreciate about Newton Protocol is that it's trying to address a problem that actually exists instead of inventing one just to justify a token. Secure automation is a real challenge. Giving software limited authority instead of unlimited authority makes sense. If AI is going to become part of crypto infrastructure, then permission systems probably deserve much more attention than flashy demos or promotional videos.
But none of that guarantees success. Maybe it works, maybe it doesn't. Crypto has a habit of making the obvious look difficult and the impossible look easy until reality catches up. Plenty of sensible ideas have failed because they couldn't attract users. Plenty of questionable ideas have succeeded simply because they captured attention at the right moment. That's one of the strange things about this industry. Merit and market performance don't always travel together.
After spending enough years watching this space, I've become much less interested in promises and much more interested in behavior. I want to see whether people actually trust these systems with real money. I want to see whether developers build on them after the initial excitement disappears. I want to see whether the project still feels relevant once the AI narrative inevitably cools down and crypto starts chasing the next big thing.
So I don't look at Newton Protocol with excitement, and I don't look at it with complete dismissal either. I look at it with cautious curiosity. The problem it's trying to solve feels genuine, which already puts it ahead of plenty of projects I've seen over the years. Whether it becomes important depends less on the quality of its story and more on whether it quietly earns trust over time.
Crypto has taught me that the loudest projects aren't always the ones that matter, and the most useful infrastructure often gets ignored until people suddenly realize they needed it all along. Maybe Newton Protocol ends up being one of those quiet success stories. Maybe it fades into the long list of ambitious experiments that never found enough users. Right now, I honestly don't know, and after everything this industry has put investors through, I'm perfectly comfortable admitting that.
@NewtonProtocol #Newt $NEWT

Il arrive un moment où passer suffisamment d’années dans la crypto change la façon dont vous voyez chaque nouveau projet. Au début, tout semble révolutionnaire. Chaque livre blanc promet un nouveau système financier, chaque influenceur a un « diamant caché », et chaque token affirme construire l’avenir. Après quelques krachs de marché, des pannes d’échange, des feuilles de route abandonnées et des communautés qui disparaissent du jour au lendemain, cette excitation finit lentement par laisser place à la prudence. Honnêtement, ces jours-ci, ma première réaction à n’importe quel nouveau projet crypto n’est pas la curiosité. C’est le scepticisme.

Honnêtement, je ne sais pas combien de narratifs crypto je suis censé trouver excitants en plus.
On dirait qu’à chaque fois qu’il s’écoule quelques mois, l’industrie trouve un nouveau mot à la mode, l’emballe dans un token et dit à tout le monde que c’est l’avenir. D’abord les ICO. Ensuite la DeFi. Puis les NFT. Ensuite le métaverse. Maintenant, tout est de l’IA. Apparemment, chaque projet a une couche IA, des agents IA, une infrastructure IA, une automatisation IA, de l’IA ceci et de l’IA cela. À un moment donné, il est devenu difficile de distinguer les idées véritablement intéressantes du marketing créatif.

Je ne sais pas si c’est juste moi qui vieillis dans les cryptos, mais à un moment donné, en chemin, j’ai cessé de m’exciter à chaque fois qu’un nouveau token apparaissait. À une époque, chaque lancement ressemblait au début de quelque chose de gigantesque. Chaque livre blanc promettait une révolution financière. Chaque influenceur, d’une manière ou d’une autre, parvenait à découvrir « la prochaine grande chose » trois fois par semaine. Chaque feuille de route avait l’air de science-fiction écrite par quelqu’un qui n’avait jamais livré de produit auparavant.
Après suffisamment de cycles de marché, toutefois, l’excitation finit par s’user.