NEAR is making a selected wager: that the way forward for crypto is autonomous AI brokers transacting at machine pace, and that they are going to want a blockchain constructed to deal with them. A June improve is the centerpiece. Right here is the thesis, the expertise, and the one quantity that complicates it.
Abstract
- NEAR is betting that AI brokers will want a blockchain constructed for machine-speed transactions.
- Dynamic resharding is the June improve designed to scale capability routinely.
- NEAR Intents offers brokers a technique to settle exercise throughout a number of chains.
- The thesis is coherent, however falling lively customers present the agent economic system has not arrived but.
NEAR Protocol has spent 2026 rebuilding its pitch to the market round a single, particular guess: that the way forward for crypto belongs to autonomous AI brokers, software program that transacts by itself at machine pace, and that these brokers will want a blockchain engineered to deal with them.
The wager is sharp and weird in a subject filled with imprecise AI branding, as a result of NEAR is pointing at a concrete use case, an on-chain economic system the place AI brokers purchase compute, settle funds, label knowledge, and execute trades routinely. These brokers might generate bursts of transactions that might paralyze a standard blockchain, and NEAR is positioning itself because the infrastructure constructed to soak up that load.
A serious community improve in June 2026, introducing automated scaling, is the centerpiece of the guess, and NEAR’s management has branded the token “the currency of agents” and the community “a unified commerce layer.” The thesis is deeply fascinating, the expertise is actual, and there may be one quantity that complicates the entire story.
This piece works by NEAR’s guess in full: the AI-agent thesis and why a blockchain for brokers would should be totally different, the June improve referred to as dynamic resharding and what it truly does, the opposite items NEAR has assembled across the thesis together with its cross-chain settlement system and its privateness tooling, the tokenomics that tie utilization to the token’s worth, and the trustworthy complication, a spot between NEAR’s hovering narrative and its precise on-chain utilization that each critical observer ought to weigh.
The objective is to elucidate what NEAR is attempting to grow to be and to evaluate the guess clearly, neither dismissing an actual and bold technical effort nor accepting the narrative uncritically. NEAR is without doubt one of the extra concrete expressions of the AI-crypto thesis, and understanding it illuminates the place that entire concept stands.
The guess: a blockchain constructed for AI brokers
To grasp NEAR’s technique, it’s important to perceive the precise future it’s betting on, as a result of the complete technical effort follows from a specific imaginative and prescient of how crypto might be used.
The imaginative and prescient is an on-chain economic system populated by autonomous AI brokers, software program applications that act on their very own to perform targets, transacting with one another and with providers at machine pace and scale. On this future, an AI agent may want to purchase computing energy on one blockchain, settle a fee on one other, and retailer knowledge on a 3rd, all routinely.
A swarm of such brokers reacting to a possibility, a worthwhile arbitrage, or a big data-labeling job, might out of the blue generate a whole bunch of 1000’s of transactions in a brief span. It is a basically totally different utilization sample from human-driven crypto, the place transactions arrive at human tempo and human scale.
Brokers function at machine frequency, in unpredictable bursts, and at volumes that might overwhelm a blockchain designed for human customers, which is the issue NEAR has determined to unravel. That makes it one other AI-crypto crossover, however one centered on transaction infrastructure moderately than id.
NEAR’s co-founder, who notably co-authored the 2017 analysis paper that launched the transformer structure underlying right this moment’s giant language fashions, has framed the protocol as elementary infrastructure for precisely this AI-driven commerce.
Why would AI brokers want a unique blockchain as a substitute of utilizing current ones? The reply is about dealing with unpredictable, machine-speed demand with out breaking.
On a standard blockchain, a sudden explosion of transactions causes congestion: charges spike, confirmations gradual, and the community turns into costly and sluggish for everybody. That’s deadly for AI brokers that must transact cheaply, immediately, and at scale with out warning.
A blockchain serving AI brokers should be capable of take in sudden, huge surges of exercise whereas protecting charges low and confirmations quick, scaling up its capability routinely the second demand spikes. There is no such thing as a time to attend for human intervention when a swarm of brokers begins transacting.
This requirement, automated, immediate scalability to deal with unpredictable machine-speed bursts, is the technical coronary heart of NEAR’s guess, and it’s what the June improve is designed to ship. NEAR is wagering that whoever builds the blockchain that may deal with AI brokers at scale will grow to be important infrastructure for the agent economic system, and it’s attempting to be that blockchain.
The June improve: dynamic resharding
The centerpiece of NEAR’s guess is a June 2026 improve referred to as dynamic resharding, and understanding what it does, in plain phrases, explains why NEAR thinks it could serve AI brokers when different blockchains can not.
The idea rests on sharding, a method NEAR has used since its launch to scale its blockchain. Sharding splits a blockchain into a number of parallel partitions referred to as shards, every processing transactions independently, like opening a number of checkout strains in a grocery retailer as a substitute of forcing everybody by a single queue.
Extra shards imply extra transactions processed in parallel, and subsequently extra capability. For a primary primer on the ledger mannequin, sharding and scaling defined begins with the blockchain construction that sharding modifies.
NEAR has scaled this manner for years, however till now, including a brand new shard was a gradual, handbook course of requiring weeks of validator coordination, a governance vote, and a staged rollout. That’s the equal of needing a committee assembly each time the shop needed to open one other checkout line.
That handbook bottleneck is strictly the issue for AI brokers, as a result of when a surge of agent exercise hits, there isn’t a time to convene a vote and coordinate validators over weeks. The capability has to seem instantly or the community congests.
Dynamic resharding removes the human bottleneck solely. With the improve, when a shard fills up previous an outlined threshold, it routinely splits into extra shards, deterministically and with none human intervention, including capability in actual time precisely when and the place it’s wanted.
Within the grocery-store analogy, the shop now routinely opens new checkout strains the second the prevailing ones get crowded, with no supervisor required. NEAR’s management says the improve will let the community scale to many dozens of shards, with throughput exceeding that of main fee networks.
They body it as foundational to the AI-agent imaginative and prescient: when a swarm of brokers out of the blue floods the community, dynamic resharding isolates that surge into newly created shards, absorbing it whereas protecting charges flat and confirmations quick for everybody else.
The identical improve additionally provides post-quantum-secure signatures, cryptographic protections designed to withstand future quantum computer systems, letting customers rotate to quantum-safe keys. That could be a forward-looking safety measure that alerts NEAR’s ambition to be sturdy infrastructure.
The improve, a part of NEAR’s community launch numbered 2.13, is the technical supply of the AI-agent guess: automated, immediate scaling constructed exactly for the unpredictable machine-speed demand that brokers would generate.
The items across the guess
Dynamic resharding is the centerpiece, however NEAR has assembled a number of different items across the AI-agent thesis, and seeing them collectively exhibits that the guess is a coordinated technique, not a single characteristic.
Crucial supporting piece is NEAR’s cross-chain settlement system, referred to as Intents, which addresses an issue particular to AI brokers working throughout a number of blockchains. Slightly than requiring an agent to carry tokens on each chain and navigate the complexity of shifting between them, the Intents system lets an agent merely categorical what it needs to perform, and specialised members referred to as solvers work out the optimum path throughout chains to make it occur.
For an AI agent that should purchase compute on one chain, choose one other, and retailer knowledge on a 3rd, this abstraction is strictly what makes working throughout a fragmented multi-chain world sensible. The Intents system has processed a big quantity of cross-chain exercise, producing tens of thousands and thousands of {dollars} in charges and settling transactions throughout many dozens of blockchains.
It’s central to NEAR’s pitch as a “unified commerce layer” for brokers, the connective tissue that lets brokers transact throughout the entire crypto ecosystem by one interface.
NEAR has additionally leaned closely into privateness, the second supporting pillar, on the reasoning that AI-driven commerce and confidential finance require privateness ensures. The protocol’s infrastructure powers merchandise providing confidential on-chain treasuries, personal multisig, payroll, and stability administration for organizations that must handle funds with out exposing every thing publicly.
Individually, NEAR’s AI division rolled out automated anonymization of private info in prompts despatched to closed AI fashions, scrubbing delicate knowledge earlier than it reaches the inference infrastructure. That addresses enterprise issues about knowledge leakage when utilizing AI.
Along with dynamic resharding, these items, cross-chain settlement by Intents and a set of privateness instruments, type a coordinated thesis: NEAR is attempting to be the scalable, cross-chain, privacy-capable settlement layer that AI brokers and confidential finance want. It’s assembling the precise capabilities that an agent economic system would require as a substitute of simply including a generic AI label.
That technique is coherent, which is a part of what makes the guess credible sufficient to take severely.
The tokenomics: tying utilization to worth
For buyers, the query is how NEAR’s technical ambitions connect with the worth of the NEAR token, and the protocol has restructured its tokenomics to forge that hyperlink, which is price understanding.
NEAR made two essential tokenomic modifications that tie community utilization to token worth. First, it reduce its inflation charge, decreasing the utmost annual issuance of latest tokens considerably, which issues as a result of the token provide is now totally unlocked, so decrease issuance means much less dilution of current holders.
Second, and extra instantly tied to the AI-agent thesis, NEAR activated a charge mechanism on its Intents settlement system, below which the charges generated by cross-chain settlement exercise are used to purchase NEAR tokens on the open market. This creates a direct suggestions loop: extra utilization of the Intents system generates extra charges, and people charges translate into extra shopping for stress on the token.
Meaning if AI-agent and cross-chain exercise grows, the expansion flows by to demand for NEAR. The design is supposed to make sure that the token captures worth from the community’s precise utilization as a substitute of relying purely on hypothesis, aligning the token’s worth with the success of the AI-agent thesis.
The proof-of-stake base issues too, as a result of staking is how networks like NEAR safe themselves whereas issuing rewards and aligning validators. That’s NEAR’s proof-of-stake basis, and it sits beneath the scaling and utilization story.
This tokenomic construction is what makes the AI-agent guess an funding thesis and never only a technical one. If NEAR succeeds in changing into the settlement layer for AI brokers, the ensuing surge in transaction exercise would generate charges that purchase NEAR, and the lowered inflation would imply that demand is just not offset by heavy new issuance.
The logic is clear: utilization drives charges, charges drive token shopping for, and decrease inflation preserves the impact, so the token is engineered to profit if the agent economic system materializes on NEAR. The caveat, which the subsequent part develops, is that this complete mechanism is dependent upon actual, rising utilization.
The fee-to-buyback loop solely generates significant demand if the Intents system and the broader community are literally getting used at scale. A intelligent tokenomic design that ties worth to utilization is just as worthwhile because the utilization it captures, and that’s exactly the place NEAR’s story meets its complication.
The construction rewards success, but it surely can not manufacture it.
The quantity that complicates the story
Right here is the trustworthy complication that any critical evaluation of NEAR should confront, as a result of it’s the hole between the narrative and the truth, and it’s the single most essential factor for a skeptical observer to weigh.
NEAR’s token has rallied considerably on the AI-agent thesis, surging on the announcement of dynamic resharding and the broader AI narrative. The story is compelling, the expertise actual, the technique coherent.
However the on-chain utilization tells a extra sobering story. The variety of day by day lively customers on the NEAR community fell dramatically over 2026, dropping from almost three million earlier within the yr to a small fraction of that, a steep decline that analysts have flagged as a warning signal exactly as a result of it diverges so sharply from the hovering worth and narrative.
That is the hole that complicates every thing: NEAR’s worth and story level to a thriving AI-agent future, whereas its precise utilization, measured by lively customers, has been falling, not rising. The narrative describes a community about to be flooded with AI-agent exercise, whereas the information exhibits fewer people truly utilizing it.
That disconnect between worth motion and on-chain utilization is strictly the form of sign that ought to make an observer cautious.
This doesn’t imply the guess is doomed, but it surely means the guess is unproven and largely forward of its proof. A few of NEAR’s rally has been pushed by elements apart from elementary adoption, together with quick squeezes that power bearish merchants to purchase again positions and amplify upward strikes, and by the highly effective pull of the AI narrative itself, which may raise a token’s worth quicker than actual utilization justifies.
The essential open query is whether or not the AI-agent thesis will translate into precise, sustained on-chain exercise: whether or not the charges, the utilization, the agent transactions, and the income seize will genuinely develop sufficient to justify the renewed market consideration and the token’s worth.
The expertise may match as marketed and the technique could also be sound, however the agent economic system NEAR is betting on has not but arrived at scale on its community. The falling consumer depend is a reminder that the thesis stays a wager on the long run, not an outline of the current.
An trustworthy evaluation holds each truths: NEAR has constructed coherent, fascinating infrastructure for a believable future, and that future has not proven up within the utilization knowledge but, leaving the guess credible however unproven.
The right way to weigh the guess
For anybody attempting to evaluate NEAR, the state of affairs comes all the way down to weighing an actual and coherent technical guess towards an unproven thesis and a worrying utilization pattern, and some rules make clear the judgment.
The case for taking NEAR severely is actual. The AI-agent thesis is believable, a way forward for autonomous brokers transacting on-chain is a reputable path for crypto, and NEAR has constructed a coherent, technically formidable set of instruments for it: automated scaling by dynamic resharding, cross-chain settlement by Intents, privateness infrastructure, and tokenomics that tie utilization to token worth.
This isn’t imprecise AI branding bolted onto an unrelated chain; it’s a centered, multi-year effort to construct particularly for the agent economic system, led by a workforce with deep AI credentials. If the AI-agent future arrives and NEAR captures even a significant share of it, the community’s design positions it to profit considerably, and the tokenomics would channel that profit to the token.
For an investor who believes within the AI-agent thesis and in NEAR’s execution, the guess has a transparent logic. Brokers would use the code brokers would transact by, and NEAR is attempting to make that code scale throughout chains and bursts of exercise.
The case for warning is equally actual and rests on the hole between narrative and actuality. The thesis is unproven, the agent economic system has not arrived at scale, the on-chain utilization has been falling, not rising, and a part of the worth power has come from market mechanics like quick squeezes and the momentum of the AI narrative as a substitute of from elementary adoption.
An investor ought to weigh that the guess is exactly that, a guess on a future which will or might not materialize on NEAR particularly, in a aggressive subject the place different blockchains are additionally pursuing scalability and AI use instances. Computerized scaling, if it proves worthwhile, may very well be matched by rivals.
The disciplined studying is to deal with NEAR as a high-conviction guess on a selected and unproven future, sized to the truth that the thesis is forward of the proof. Watch the precise utilization knowledge, the charges, the lively customers, and the agent exercise, as a result of these are the true exams of whether or not the narrative is changing into actuality.
That self-discipline issues particularly towards the market backdrop for altcoins, the place robust narratives can nonetheless run right into a troublesome macro and liquidity surroundings. The expertise and technique are actual; the adoption is the open query, and watching it, not the worth, is easy methods to know whether or not the guess is paying off.
None of that is funding recommendation; it’s a body for assessing considered one of crypto’s extra concrete and bold AI bets with applicable readability about what’s confirmed and what’s hoped.
A coherent guess, forward of its proof
NEAR’s wager is without doubt one of the clearest expressions of the AI-crypto thesis available in the market: a guess that autonomous AI brokers will transact on-chain at machine pace and scale, and that they are going to want a blockchain constructed to soak up that load.
The June dynamic resharding improve is the centerpiece, delivering automated, immediate scaling designed exactly for the unpredictable bursts an agent economic system would generate. Round it, NEAR has assembled a coherent technique: cross-chain settlement by Intents, privateness tooling, and tokenomics that channel usage-driven charges into shopping for the token.
Led by a workforce with deep AI credentials and pointed at a believable future, the guess is particular, technically actual, and value taking severely, not the imprecise AI branding that decorates so many crypto tasks.
The complication is the hole between the narrative and the proof. NEAR’s worth and story describe a community on the verge of an AI-agent increase, whereas its precise utilization, measured by a day by day lively consumer depend that has fallen sharply over 2026, tells a extra sobering story.
A part of the rally has come from quick squeezes and the pull of the AI narrative, not elementary adoption. The agent economic system NEAR is betting on has not but arrived at scale on its community, which leaves the thesis credible however unproven, forward of its proof.
The trustworthy evaluation holds each: NEAR has constructed spectacular, centered infrastructure for a plausible future, and that future has not proven up within the utilization knowledge, making NEAR a high-conviction guess on a selected future, not an outline of current actuality.
Whether or not dynamic resharding and the Intents system grow to be the rails of an actual agent economic system, or whether or not the narrative outruns the adoption, is the query that may outline NEAR. The reply lies not within the worth however in whether or not the brokers ever truly arrive.
The guess is positioned and the infrastructure is constructed; the economic system it’s constructed for has but to point out up.
Continuously requested questions
What’s NEAR betting on with the AI-agent thesis?
NEAR is betting that the way forward for crypto includes autonomous AI brokers, software program that transacts by itself at machine pace, and that these brokers will want a blockchain engineered to deal with their unpredictable, high-volume exercise. It envisions an on-chain economic system the place brokers purchase compute, settle funds, and retailer knowledge routinely, producing bursts of transactions that might overwhelm typical blockchains. NEAR is positioning itself because the scalable settlement layer for this agent economic system, branding its token “the currency of agents.”
What’s dynamic resharding?
Dynamic resharding is a June 2026 NEAR improve, a part of community launch 2.13, that lets the blockchain routinely add capability when demand spikes. NEAR makes use of sharding, splitting the community into parallel partitions, or shards, like a number of checkout strains. Beforehand, including a shard required weeks of handbook validator coordination and a governance vote. Dynamic resharding removes that bottleneck: when a shard fills up, it routinely splits into extra shards, with no human intervention, including capability in actual time, which is crucial for absorbing sudden AI-agent surges.
Why would AI brokers want a particular blockchain?
As a result of they transact at machine pace in unpredictable bursts. A swarm of brokers reacting to a possibility might generate a whole bunch of 1000’s of transactions out of the blue, which on a standard blockchain causes congestion, spiking charges and slowing confirmations for everybody. A blockchain serving brokers should take in these surges routinely whereas protecting charges low and confirmations quick, scaling capability the moment demand spikes, as a result of there isn’t a time for human intervention. That automated, immediate scalability is what NEAR’s dynamic resharding is constructed to supply.
How does NEAR’s token seize worth from this?
By means of two tokenomic modifications. NEAR reduce its inflation charge, decreasing dilution because the provide is totally unlocked. Extra importantly, it activated a charge mechanism on its Intents cross-chain settlement system, the place charges from settlement exercise are used to purchase NEAR on the open market. This creates a suggestions loop: extra utilization generates extra charges, which purchase extra NEAR, so development in AI-agent and cross-chain exercise flows by to token demand. The design ties the token’s worth to precise community utilization moderately than pure hypothesis.
What’s the drawback with NEAR’s story?
A spot between narrative and actuality. NEAR’s worth and story describe a thriving AI-agent future, however its on-chain utilization tells a unique story: day by day lively customers fell sharply over 2026, from almost three million to a small fraction of that. This decline diverges from the hovering worth, and analysts flag it as a warning signal. A part of the rally additionally got here from quick squeezes and AI-narrative momentum moderately than elementary adoption. The thesis is unproven, and the agent economic system has not but arrived at scale on NEAR.
Is NEAR a great funding?
That is dependent upon whether or not you imagine the AI-agent thesis and NEAR’s execution, and it’s genuinely unproven. The case for it: a believable future, coherent and bold expertise, a credentialed workforce, and tokenomics tying worth to utilization. The case for warning: the thesis is unproven, utilization has been falling, the agent economic system has not materialized at scale, rivals are pursuing related targets, and worth power has partly come from market mechanics. The disciplined strategy watches precise utilization knowledge, not worth, because the check. This isn’t funding recommendation.
As of June 21, 2026. Crypto markets and protocol particulars change rapidly; confirm present knowledge earlier than counting on this evaluation. This text is info, not funding recommendation.
