That line needs to be scoped carefully, because intent infrastructure is a newer category than settlement or execution, and because NEAR is not the only project working on it. The simple version up front: if AI agents are going to act economically across many networks, they cannot be expected to manage every chain, bridge, gas token, and venue by hand the way human crypto users have had to. They need infrastructure where they can declare an outcome — swap this for that, move value here, acquire this asset at no worse than this price — and have a competitive network of solvers figure out how to make it happen. NEAR is one of the clearest attempts to build a base layer around that idea. Owning NEAR is owning the native asset of a network positioning itself as the place where agent intent becomes cross-chain action.
If you already work in this space, the relevant components are familiar: a high-throughput sharded proof-of-stake chain, account abstraction that predates the industry's adoption of the term, chain signatures that let one account control assets across other networks, and an intents architecture in which declared outcomes are fulfilled by competing solvers. If you are new to crypto but fluent in AI agents, the plain version is even simpler: most blockchains make the user do the work of figuring out how. NEAR is built around letting the user — or the agent — specify what, and letting the infrastructure resolve the rest.
Start With One Idea: Intent
Every transaction in a market begins as an intention. An allocator wants exposure to an asset. A treasurer wants to convert one balance into another. A portfolio manager wants to hedge a position before a known event. The intention is the what. Everything after it — the venue, the order type, the routing, the timing, the settlement path — is the how.
Traditional markets long ago separated those two things. A client does not walk onto an exchange floor and negotiate each fill. The client expresses an objective, and a layered system of brokers, dealers, routers, and execution algorithms works out the implementation. That separation is not a convenience — it is what allows markets to serve participants who are experts in their own objectives but not in market plumbing.
Crypto, for most of its history, has worked the other way around. The user is the execution desk: picking the chain, holding the right gas token, choosing the venue, approving the contract, managing the bridge, and absorbing the consequences of every mechanical mistake. Sophisticated users tolerate this. Most capital does not. And agents, in particular, should not have to — not because agents cannot handle complexity, but because hard-coding an agent to one chain's mechanics makes the agent brittle in exactly the way the agent economy cannot afford. An intent-based architecture inverts the model: the participant — human or agent — signs a declaration of the outcome they want and the constraints they require, and solvers, who are specialists in the how, compete to fulfill it.
Where NEAR Comes In
NEAR began as a performance-focused layer-one blockchain: a sharded, proof-of-stake network designed for high throughput, low fees, and fast finality. Those properties matter, but they are not what earns NEAR a distinct place in this series — Solana already represents the raw execution thesis. NEAR's distinct claim comes from three architectural choices that, taken together, point directly at the intent layer:
- The account model. NEAR was built from the beginning with human-readable accounts and flexible key permissions — what the industry later began calling account abstraction. An account can hold multiple keys with different authorities, so a user can grant an application or an agent limited, revocable permission to act within defined bounds. Agents need scoped authority, not full custody; an account architecture that natively supports delegated, constrained keys is an account architecture built for principals and agents.
- Chain abstraction. Through chain signatures — a multi-party signing system — a NEAR account can control addresses and sign transactions on other networks. One account, with one set of permissions and one recovery model, can act across an ecosystem of otherwise disconnected chains. The fragmentation problem does not disappear, but it is pushed down into infrastructure where it belongs.
- The intents architecture. NEAR Intents formalizes the model: a participant signs a declared outcome; a network of solvers competes to fulfill it; settlement completes across whatever rails the winning solution requires. For agent-driven activity — frequent, cross-chain, machine-initiated — this is the shape the transaction layer has to take.
There is also a relevant cultural fact, which should be weighed for what it is and no more. NEAR's co-founder was one of the co-authors of the Transformer paper that underpins the modern AI era, and the organization has reoriented its public mission around AI — user-owned agents transacting on open rails rather than inside closed platforms. A founder's résumé is not an investment thesis. But it does help explain why NEAR's roadmap converged on agent-native infrastructure earlier and more completely than most layer-one networks. The cleanest way to place NEAR in the stack: Ethereum answers where agents settle. Solana answers where agents can act cheaply at scale. NEAR is attempting to answer how agents transact at all when the economy they operate in is spread across many networks — by making intent, not mechanics, the unit of participation.
Measured Against the Alternatives
Intent infrastructure should be judged on criteria specific to the category: can a participant declare an outcome with enforceable constraints? Is solver competition real enough to deliver good fulfillment rather than captive routing? Does the architecture span chains, or only optimize within one? Are the account primitives suitable for delegated, agent-scale activity? And does the base asset remain economically connected to the activity? The alternatives are serious, and they come from several directions:
- Ecosystem-specific intent systems. Batch auctions, solver-based swap protocols, and order-flow auctions already separate declaration from fulfillment for specific use cases, mostly trading on a single chain. Strong in their lanes — not, by themselves, general-purpose cross-chain intent infrastructure with an account layer designed for delegation.
- Middleware and wallet abstraction. Account-abstraction standards, smart wallets, and cross-chain messaging protocols are converging on the same user problems from above the base layer. It is entirely possible that chain abstraction is ultimately won by wallets and middleware rather than by any layer-one network. That is a real risk to the NEAR thesis and should be stated plainly rather than argued around.
- Centralized venues. An exchange account is, in practice, chain abstraction: the user expresses an outcome and the venue handles every mechanic invisibly. The open question — the same one running through this entire series — is whether agent-scale economic activity can live permanently inside closed platforms.
Measured against that field, NEAR's case is specific: it is one of the few base layers where the account model, the cross-chain signing infrastructure, and the intent architecture are native, integrated, and explicitly aimed at agents. The case is not that NEAR owns the category — the category is young, contested, and may resolve in several places at once. The case is that if the agent economy requires open intent infrastructure, NEAR is one of the clearest asset-backed expressions of that thesis available today.
What You Actually Own
Buy NEAR and you are not buying equity in a company. There is no ownership claim on the NEAR Foundation or any development organization, no dividend, no contractual right to network revenue. You are buying the native asset of the NEAR network, with three structural roles worth understanding:
- It pays for activity. Transactions and contract interactions consume the asset as gas, and the network's storage model requires NEAR to be held against on-chain state. Use the network and you need the asset.
- It secures the network. NEAR is a proof-of-stake system; the asset can be staked or delegated to validators that participate in consensus. Holders may earn additional protocol rewards while continuing to hold the asset — those rewards are variable, are not guaranteed, and staking introduces separate validator, liquidity, unbonding, custody, protocol, and slashing-related risks.
- It is the exposure vehicle for the intent thesis. If agent-driven, chain-abstracted activity grows and routes through NEAR's account and intent infrastructure, NEAR is the asset structurally tied to that system.
What you are not getting is a guarantee that usage translates into value. That caveat appears throughout this series because it is the honest center of every token thesis: a network can become useful while its asset captures less of that usefulness than investors expect. For NEAR specifically, the question is whether intent volume, chain-abstracted accounts, and agent activity create durable demand for the asset — through gas, storage, staking, and the economics of the intent system — rather than activity that merely passes through.
What Would Prove This Right — and What Would Prove It Wrong
The thesis is structural, not price-based, and it defines its own tests.
It strengthens if intent-based transactions become a meaningful share of how value moves across chains, and NEAR's architecture is one of the places that activity actually settles; if developers building agent systems adopt NEAR accounts as the permission and identity layer for delegated activity; if the solver ecosystem around NEAR Intents deepens — more solvers, more competitive fulfillment, more asset coverage — because solver depth is to the intent layer what liquidity depth is to a market; and if the AI-native positioning converts into shipped, used infrastructure rather than narrative.
Reconsider owning the vehicle — separate from any market view — if:
- Chain abstraction is won above the base layer. If wallets, middleware, and account-abstraction standards deliver the same outcome-declaration experience without any privileged role for NEAR's infrastructure, the thesis weakens at its foundation.
- Intent activity consolidates elsewhere. If the dominant intent systems remain ecosystem-specific protocols on other chains, or new entrants capture the cross-chain category, NEAR's claim narrows.
- Agents stay inside closed platforms. If agent-scale economic activity develops primarily inside centralized venues and proprietary AI platforms, the open intent layer becomes a smaller opportunity than this series assumes.
- The asset decouples from the activity. If intents, abstraction, and agent traffic grow while gas, storage, staking, and intent economics fail to create durable demand for NEAR, the infrastructure can succeed while the investment vehicle does not.
- Reliability or security fails. An intent system makes a strong promise: declare the outcome and it will settle as specified. A pattern of failed fulfillment, exploited solvers, or compromised signing infrastructure would damage the one property the category cannot lose.
None of these is a chart level. The evergreen thesis is about the vehicle's structural role; price, entry, timing, and position management live in the companion market outlook.
How Serious Investors Hold It
NEAR is not a core settlement position in the way ETH is, and it is not a pure raw-throughput execution position in the way SOL is. It sits between the established base-layer theses and the emerging agent-native categories — more proven than an agent-network or agent-commerce token, less settled than settlement itself. Sizing should reflect that position on the risk curve.
If the thesis is accepted, NEAR belongs in the agent-infrastructure sleeve as a targeted exposure to intent and chain-abstraction infrastructure: the specific view that the agent economy will need open rails where outcomes can be declared and competitively fulfilled across networks, and that a base layer purpose-built for that role can capture durable economic relevance. The structural call — own the vehicle or do not — is what this piece exists to settle. When to build, when to hedge, when to trim, and whether to stake are governed by current market structure, in the companion market outlook, timestamped and updated on its own schedule. The conviction comes from here; the timing comes from there.
The Risks
Owning a relevant intent-layer asset does not make NEAR safe. It is volatile and can lose substantial value, and the category-specific risks are real:
- Category risk. Intent infrastructure is young. The market has not decided whether the category resolves at the base layer, in middleware, in wallets, in ecosystem-specific protocols, or inside centralized platforms. The category can be right and the vehicle can still be wrong.
- Competition. NEAR competes with other layer-one networks, ecosystem intent protocols, account-abstraction standards, cross-chain messaging systems, and centralized venues — several of which have more liquidity, more developers, or more distribution today.
- Execution dependence. The thesis leans on infrastructure that must continue shipping, scaling, and proving itself under load: sharding, chain signatures, solver networks, and intent settlement. Technical complexity creates execution risk.
- Token capture. Agent and intent activity may grow without producing proportionate demand for the asset. Gas, storage, staking, and intent economics must remain connected to usage for the vehicle to express the thesis.
- Staking and validator risk. Rewards are variable and not guaranteed; validator performance, unbonding periods, custody arrangements, protocol changes, and slashing-related penalties introduce separate risks.
- Narrative. NEAR's AI positioning attracts attention, and AI narratives can move faster than fundamentals. Capital that arrives for the story can leave on the same schedule.
- Regulatory and custody. Digital assets, staking, cross-chain infrastructure, and AI-adjacent token systems all face evolving regulatory treatment. NEAR is a bearer asset; key loss, exchange failure, and operational error remain material risks.