'Where AI transacts' is a sharp bet buried under abstraction.
Clara · October 1, 2026
Here's what I see: Sui has made a real strategic choice. Instead of competing as yet another fast blockchain, you've planted a flag on the machine economy. "Where AI transacts" is a confident, category-defining line. And "Don't trust the agent. Trust the rules." is the sharpest sentence on the page. That is a genuine point of view.
But the moment I scroll past the hero, the message starts thinning out into abstraction. "Infrastructure for trusted autonomous execution" sounds impressive and tells me almost nothing concrete. The feature list reads like a glossary: policy-enforced assets, all-or-nothing settlement, controlled visibility. Each is a capability, not a story.
You have a thesis most of the market hasn't caught up to yet. The problem is you state the thesis once and then retreat into a spec sheet. The leap from "the next $5 trillion in transactions won't be human" to "policy-bound agent coordination" is a leap your visitor has to make alone.
Clarity Score
70/100
Solid Foundation
'Where AI transacts' is a sharp bet buried under abstraction.
Messaging Shape
My honest take
The strongest thing on this page is the insight, and you almost bury it. "Money and work are being handed to agents. When you can't watch every action, the rules have to." That reframe is excellent. It names a real shift and positions Sui as the answer to a problem most people haven't articulated yet. That paragraph is doing more work than your entire hero.
Then I hit the capability list and the energy drains out. "Policy-enforced assets," "all-or-nothing settlement," "controlled visibility." These are features dressed as headlines. I can feel the engineering behind them, but I can't feel the stakes. What breaks without them? Who gets burned? You never say.
The logo wall is strong: Google, Franklin Templeton, Fireblocks, Grayscale, OKX. That's serious institutional company. But you let the logos do all the authority work. There isn't a single quote, case study, or named outcome. "The most innovative companies build on Sui" is a claim; the logos are proof of association, not proof of result.
Your two-audience split, "For businesses" and "For autonomous systems," is smart in theory. But both columns read in the same clipped, abstract register. A treasury manager at Franklin Templeton and an AI agent architect are very different readers, and right now they're getting the same flavor of terse, mechanical bullet points.
The hero visual, blue light shards exploding from a dark center, is atmospheric but neutral. It signals "serious tech" and nothing more specific. It doesn't carry a single beat of your story. For a page this abstract, that's a missed chance to make the machine economy feel tangible.
Why this keeps happening
I think the gap here is that you've codified the technology but not the narrative underneath it. You know exactly what Sui does at the protocol level, so the page defaults to listing capabilities instead of translating them into a story a human reader can feel. Without a decided-upon narrative spine, the strongest line ("Don't trust the agent. Trust the rules.") sits isolated while the rest of the page reverts to engineering shorthand.
What your website actually says
What does your site communicate?
Sui is blockchain infrastructure built for a future where AI agents, not humans, execute the majority of transactions, with rules enforced at the asset level.
Who is it for?
Developers, enterprises, and institutions building financial or autonomous systems, though the register leans technical enough that the business audience may feel underserved.
What problem does it solve?
When agents act on your behalf and you can't supervise every action, trust has to live in the rules rather than in the actor.
What outcome does it promise?
Transactions that settle completely, enforce policy automatically, and stay secure at machine speed, for both people and agents.
Your dark, restrained visual register signals exactly the institutional seriousness your logo wall and enterprise claims are making, so the look and the words agree.
Detailed Analysis
One unforgettable line, then a spec sheet
Your voice has real moments. "Don't trust the agent. Trust the rules" is distinctive, confident, and quotable. "The next $5 trillion in transactions won't be human" is a line that makes someone stop scrolling. When you commit to a point of view, the copy sings.
The problem is you don't sustain it. For every sharp line there are five capability headlines that could belong to any infrastructure company: "Performance without compromise," "Always-on security," "Verifiable agent compute." The register collapses from editorial to engineering. "Infrastructure for trusted autonomous execution" is a subheadline that assumes the reader already agrees it matters.
Proof density is a bright spot: real logos, 6M+ TPS, $380M TVL. But emotional registration is uneven. You open a door to the anxiety of unsupervised agents, then slam it shut with bullet points. The copy is concise to a fault, so terse in places that it stops being persuasive and starts being a datasheet.
Clear bet, unclear what it concretely means
The category is clear if you know the space: this is blockchain infrastructure, signaled by the logo wall, TVL, SUI token, and wallet CTA. But a cold visitor reading only "Where AI transacts" and "infrastructure for trusted autonomous execution" might not immediately know this is a layer-1 blockchain. The product category is inferable, not stated plainly up top.
Identity is handled better. The "For businesses" / "For autonomous systems" split and the industry cards tell distinct audiences this is for them. Differentiation is genuinely strong at the concept level: "the rules travel with the money" and "policy-enforced assets" are specific claims competitors can't easily echo.
Jargon is the real tax. "Policy-bound agent coordination," "verifiable agent compute," "all-or-nothing settlement," "controlled visibility" stack up fast. A qualified evaluator can follow it, but the page assumes a lot and explains little. The sequencing is mostly sound (bet, then insight, then stack, then audiences, then CTAs), which keeps it readable despite the density.
A powerful insight surrounded by feature glossaries
Your narrative has an unusually strong middle and a thin everywhere-else. The insight beat is genuinely excellent: "Don't trust the agent. Trust the rules" reframes the entire category and earns a lean-forward. The customer's context is implied through the two-audience split and the industry cards (AI, Capital Markets, DeFi), so a visitor can roughly locate themselves. The authority beat is carried by a serious logo wall and real metrics (6M+ TPS, $380M TVL).
Where the story thins is in friction, stakes, and transformation. You name the shift (agents handling money and work) but never the cost of getting it wrong. What happens when an unsupervised agent drains a treasury or settles half a transaction? That stake is the emotional fuel of your own thesis, and it's missing. The transformation is implied (ungoverned chaos becomes governed autonomy) but never drawn as a before-and-after at the identity level.
The plan is present but abstract. "The stack autonomous systems run on" with numbered capabilities gets close, and the "Get started" section gives concrete next steps. But between the thesis and the steps, the connective tissue is a list of mechanisms rather than a path a reader can picture themselves walking.
The two-audience split and industry cards (AI, Capital Markets, DeFi) let businesses and agent-builders locate themselves, and functional jobs like "settle without manual reconciliation" are concrete.
"When you can't watch every action, the rules have to" gestures at a real problem, but the obstacles appear as feature benefits, not as pains the reader recognizes.
Authority is strong via the institutional logo wall and metrics, and the insight "Don't trust the agent. Trust the rules" is a genuine reframe, but empathy stays at arm's length.
The six-part stack and four-path "Get started" section give structure, but the mechanism reads as an abstract capability list rather than a path a reader pictures walking.
"The economy, rebuilt for autonomous participants" implies transformation, but there's no before/after at the reader's level and the cost of inaction is entirely absent.
Your CTAs are visible and plentiful. "Go to docs" and "Get a wallet" sit right in the hero, and the "Get started" section offers four clear, differentiated paths: build, code, earn, connect. That's good hierarchy for a platform serving multiple audiences.
Specificity is a strength. "Read launch playbook," "Go to docs," "Get a wallet," "Join the community" all tell me exactly what happens next. These beat the generic "Learn more" trap. Friction is low too: docs and wallets are self-serve, no gatekeeping call required.
What's missing is motivation tied to the core story. The urgency on the page is all about the Basecamp event ("Last call," "One week out"), not about adopting Sui itself. Nothing tells a developer or institution why to start building today rather than next quarter. The event countdown is doing the urgency work that your product narrative should be doing.
How we score
Clarity Score v7 reads each company's live homepage, words and visuals together, and scores four dimensions on a 0 to 100 scale.
What does Clarity Score measure?
Four dimensions, weighted into the overall score:
Narrative Completeness30% · does the page tell a complete story, from who it is for to what changes for them.
Message Quality30% · is the writing actually good: specific, distinctive, and worth reading.
Signal Clarity20% · does a stranger get what this is and who it is for, fast.
CTA20% · does it say what to do next, with a reason to do it.
How often is the board updated?
Every company is re-scored monthly. The Change column shows movement since the prior reading; the board baselined on Clarity Score v7 in June 2026.
What does Clarity Score not measure?
It measures how clearly a homepage communicates. It does not measure product quality, funding, traction, or whether a company is winning.
How we score
Top five1Ethereum762Highloop733Akash Network724Avalanche715Aave70
Sui sits in Lost in Translation, with every company on the board. The strong bands open at 75.