NEAR is asserting a transformative vision for cryptocurrency: a future where autonomous AI agents execute transactions at machine speed, requiring a blockchain tailored for such activities. A major upgrade in June is foundational to this strategy. Below, we outline the core thesis, the technology supporting it, and a key element that adds complexity to the situation.
Summary
- NEAR envisions a future where AI agents need a blockchain designed for transactions at machine speed.
- The June update introducing dynamic resharding will facilitate automatic scaling of capacity.
- NEAR Intents allows agents to operate across multiple blockchains seamlessly.
- While the premise is intriguing, a drop in active user counts indicates that the agent economy is still nascent.
By 2026, NEAR Protocol has centered its narrative on a clear, distinctive idea: that the future of cryptocurrency is intertwined with autonomous AI agents—software that can independently conduct transactions at machine speed—and that these agents necessitate a blockchain specifically designed to support their functions.
This position is sharply defined amidst a sea of vague AI marketing, as NEAR provides a concrete application: an on-chain economy where AI agents autonomously acquire computing resources, manage payments, organize data, and execute trades. These agents could generate surges in transaction volumes that may overwhelm conventional blockchains, positioning NEAR as the necessary infrastructure to meet this demand.
A significant network upgrade in June 2026 will implement automatic scalability, forming the crux of this bet. NEAR’s leadership refers to its token as “the currency of agents” and the platform as “a cohesive commerce layer.” While the underlying thesis is compelling and the technology plausible, one significant factor complicates the overall narrative.
This examination will delve into NEAR’s gamble in detail: the AI-agent thesis and the distinct needs for a blockchain suited to agents, the dynamic resharding upgrade anticipated in June and its implications, additional features NEAR has developed alongside the thesis—including its cross-chain settlement system and privacy components, the tokenomics connecting usage to the token’s value, and the challenges posed by the gap between NEAR’s ambitious narrative and its real on-chain activity, which analysts must consider.
The goal here is to clarify NEAR’s ambitions and assess the bet without dismissing an authentic and audacious technical endeavor or embracing the narrative uncritically. NEAR represents a concrete embodiment of the AI-crypto thesis, and understanding it illuminates the status of this concept.
The bet: a blockchain built for AI agents
To understand NEAR’s strategy, one must appreciate the specific future it is betting on, as this vision shapes its entire technical approach.
The vision portrays an on-chain economy consisting of autonomous AI agents, software programs that independently act to achieve goals, executing transactions with one another and with services at machine speed and volume. In this context, an AI agent might autonomously acquire computing power on one blockchain, settle payments on another, and store data on yet another.
A surge of such agents reacting to profitable opportunities or large-scale data-labeling tasks could lead to an influx of transactions in a short time frame. This contrasting usage model highlights the differences with human-driven crypto, where transactions occur at a human pace and volume.
Agents function continuously, oftentimes in unpredictable surges, and at scales that might overwhelm typical blockchains designed for human users. This is the challenge NEAR aims to address, directing its focus toward transaction infrastructure rather than user identity.
A co-founder of NEAR, who notably contributed to the 2017 paper that introduced the transformer architecture behind today’s large language models, views the protocol as essential infrastructure for this AI-centric commerce.
But why do AI agents need a specialized blockchain instead of using the existing ones? The answer lies in efficiently managing unpredictable, machine-speed demand without interruptions.
On a conventional blockchain, a sudden surge in transactions leads to congestion: fees skyrocket, confirmation times stretch, and the network becomes costly and sluggish for all users. This scenario could hinder AI agents that need to transact quickly and efficiently at scale.
A blockchain for AI agents must accommodate sudden, large-scale activity spikes while maintaining low fees and rapid confirmations, automatically scaling its capacity whenever demand surges. There’s no room for human intervention when numerous agents begin transactions.
This critical need for automatic, instantaneous scalability to manage unforeseen machine-speed demands represents the technical heart of NEAR’s bet, and it is what the June upgrade seeks to achieve. NEAR is betting that the entity that establishes a blockchain capable of supporting AI agents at scale will become an indispensable part of the agent economy, and NEAR is striving to be that entity.
The June upgrade: dynamic resharding
The essence of NEAR’s bet is a June 2026 upgrade called dynamic resharding. A straightforward understanding of its functionality clarifies why NEAR believes it can cater to AI agents where other blockchains may falter.
This concept is centered on sharding, a method NEAR has used since its inception to enhance its blockchain’s scalability. Sharding divides a blockchain into several parallel partitions, or shards, with each handling transactions independently—similar to multiple checkout lines operating simultaneously rather than forcing everyone through a single queue.
More shards translate to more transactions processed in parallel, leading to higher capacity. A foundational understanding of the ledger model begins with the blockchain structure altered through sharding.
Although NEAR has employed this scaling method for years, previously adding a shard required a slow, manual process that could take weeks of validator coordination and governance voting. This process resembles needing committee approval every time a store wishes to open another checkout line.
This manual constraint poses challenges for AI agents—when agent activity surges, there is no time to gather votes and coordinate Validators over weeks. Capacity must expand instantly or face network congestion.
Dynamic resharding eliminates the human bottleneck entirely. With this upgrade, when a shard exceeds a specific capacity threshold, it automatically divides into additional shards—swiftly and without human involvement—adding capacity in real-time precisely where and when it is needed.
In the grocery store analogy, the store autonomously opens new checkout lines the instant the existing ones become overcrowded, with no manager needed. NEAR’s leadership asserts that this upgrade will enable the network to scale to dozens of shards, allowing transaction volumes to exceed those of major payment networks.
They see it as fundamental to the AI-agent vision: as a flood of agents inundates the network, dynamic resharding will manage that influx through newly formed shards, ensuring stable fees and quick confirmations for all users.
This upgrade also enhances security with post-quantum secure signatures, cryptographic safeguards designed to withstand future quantum attacks, allowing users to transition to quantum-safe keys. This forward-thinking initiative underscores NEAR’s ambition to serve as a long-lasting infrastructure solution.
The upgrade, part of NEAR’s network release numbered 2.13, delivers the technical groundwork of the AI-agent wager: automatic, instantaneous scaling specifically catered for the unpredictable machine-speed demands agents are likely to generate.
The additional components surrounding the bet
While dynamic resharding takes center stage, NEAR has incorporated several other components that align with the AI-agent thesis. Analyzing these collectively reveals that the wager is a coordinated strategy rather than a mere standalone feature.
The most significant additional element is NEAR’s cross-chain settlement system, named Intents, which addresses a specific hurdle faced by AI agents functioning across various blockchains. Rather than requiring agents to maintain tokens on every chain and navigate the complexities of different blockchains, the Intents system empowers an agent to communicate its objectives, with specialized participants known as solvers determining the optimal pathway across chains to achieve them.
For AI agents needing to obtain computing power on one blockchain, settle on another, and store data on a third, this abstraction simplifies operating within a fragmented multi-chain environment. The Intents system has handled considerable cross-chain activity, generating millions of dollars in fees while facilitating transactions across various blockchains.
This system is pivotal to NEAR’s vision as a “unified commerce layer” for agents, serving as a connective framework that enables agents to interact across the entire crypto ecosystem from a single interface.
NEAR has also invested heavily in privacy, which serves as the second foundational pillar, operating under the premise that AI-driven commerce and confidential finance require strong privacy assurances. The protocol’s infrastructure supports initiatives that provide confidential on-chain treasuries, private multisig solutions, payroll, and fund management tools for organizations wishing to manage their finances discreetly.
Additionally, NEAR’s AI division has implemented automatic anonymization of personal data in prompts sent to proprietary AI models, cleansing sensitive information before it reaches the inference layer. This capability addresses enterprise concerns regarding data security during AI utilization.
When combined with dynamic resharding, these features—including cross-chain settlement via Intents and a suite of privacy enhancements—resonate with a unified thesis: NEAR aims to be an adaptable, privacy-enabled settlement layer meeting the needs of AI agents and confidential financial transactions. It is developing tailored capabilities essential for an agent economy rather than merely attaching a generic AI label.
This approach is coherent and solidifies the credibility behind the bet.
The tokenomics: linking usage to value
For investors, the critical question lies in how NEAR’s technical ambitions correlate with the value of the NEAR token. The protocol has reformed its tokenomics to establish this connection, which deserves scrutiny.
NEAR made two significant changes to its tokenomics intended to bridge network usage with token value. Firstly, it reduced its inflation rate, significantly lowering the maximum annual issuance of new tokens. This is crucial because the supply is now entirely unlocked, and reduced issuance helps minimize dilution for existing holders.
More importantly, NEAR introduced a fee mechanism in its Intents settlement system, whereby fees earned from cross-chain settlement activities are utilized to buy back NEAR tokens on the open market. This structure initiates a direct feedback loop: increased usage of the Intents system raises fees, which in turn generates higher demand for NEAR tokens.
Thus, if AI-agent and cross-chain actions expand, this growth will likely translate into increased demand for NEAR. The design aims to ensure that the token captures value from the network’s utility rather than from mere speculation, effectively aligning the token’s worth with the success of the AI-agent concept.
The proof-of-stake mechanism also plays an essential role, as staking secures networks like NEAR while distributing rewards and aligning validators. This proof-of-stake framework supports both the scaling and usage narratives.
This tokenomic structure transforms the AI-agent bet into a credible investment thesis rather than merely a technical proposition. If NEAR successfully positions itself as the settlement layer for AI agents, the resultant increase in transaction activity will accrue fees that bolster NEAR’s market standing, while decreased inflation alleviates pressures from new token issuances.
In essence, heightened usage generates fees, those fees drive demand for NEAR, and lower inflation preserves this dynamic, culminating in a design that benefits from the potential development of the agent economy on NEAR. However, concerns arise regarding the entire mechanism’s reliance on actual, growing usage.
The fee-to-buyback cycle fosters robust demand only if the Intents system and broader network are actively utilized at scale. An intricate tokenomic setup linking value to activity is only as effective as the genuine activity it captures, highlighting a critical juncture where NEAR’s narrative faces hurdles.
The structure incentivizes success but cannot create it.
The number that complicates the narrative
This is the genuine complication that any serious assessment of NEAR must contend with; it revolves around the disparity between the narrative and reality—an essential consideration for skeptics.
While NEAR’s token has experienced significant appreciation associated with the AI-agent narrative, soaring alongside the announcement of dynamic resharding and the broader AI narrative, the actual on-chain metrics present a more sobering view. The number of active daily users on the NEAR platform dropped dramatically throughout 2026, plummeting from nearly three million in early 2026 to a tiny fraction, a stark decrease that analysts flag as concerning given its sharp contrast to the rising price and optimistic narrative.
This gap considerably complicates the situation: while NEAR’s price and story suggest an imminent AI-agent boom, the reality of usage data reflects a decline in active participation. This disconnection between price movements and on-chain activity serves as a cautionary signal that deserves attention.
This disparity does not imply that the bet is doomed; rather, it indicates that the wager remains unproven and is largely based on anticipated developments. Some of NEAR’s price rise has derived from external factors unrelated to fundamental uptake, such as short squeezes compelling bearish traders to cover their positions, driving price movements upward, along with the enticing nature of the AI narrative itself, which can artificially inflate a token’s price more rapidly than real uptake would likely justify.
The pressing question remains whether NEAR’s AI-agent thesis will materialize into sustained and genuine on-chain activity—whether fees, usage, transactions by agents, and revenue generation will truly expand enough to validate the renewed market interest and the token’s value.
While the technology may function as intended and the strategy may be solid, the agent economy that NEAR anticipates has yet to gain traction on its platform. The declining user count serves as a reminder that the thesis exists as a future gamble rather than an accurate depiction of current circumstances.
A frank assessment must recognize both realities: NEAR has constructed a coherent and compelling framework for a plausible future, yet that anticipated future remains to be reflected in the usage data, casting NEAR as a speculative bet on a specific outcome rather than a mirror of current reality.
How to evaluate the bet
For those working to assess NEAR, the analysis culminates in balancing a substantial technical wager against an unverified thesis and concerning usage trends, with several principles clarifying the decision-making process.
The justification for taking NEAR seriously lies in its foundational reality. The AI-agent thesis is credible; the projection of autonomous agents conducting transactions on-chain is a viable direction for cryptocurrency, and NEAR has put forth a coherent, ambitious set of tools to facilitate it: automatic scalability via dynamic resharding, cross-chain settlement through Intents, privacy infrastructure, and tokenomics that connect value to usage.
This is not vague AI branding loosely affiliated with an unrelated blockchain; it embodies a dedicated, meticulous initiative oriented towards the agent economy led by a team with significant AI expertise. If the anticipated future of AI agents comes to fruition, and NEAR captures a significant share, the network’s architecture positions it to thrive, while the tokenomics would channel those benefits toward the token itself.
Conversely, the rationale for caution is equally valid and hinges on the inconsistency between narrative and reality. The thesis remains unverified, and the agent economy has yet to emerge on a substantial scale. The on-chain usage metrics indicate a decline, not growth; part of the price rise has been driven by market dynamics such as short squeezes and the allure of the AI narrative, rather than intrinsic adoption.
Investors should recognize that this bet is exactly that—a wager on a hypothetical future that may or may not materialize within NEAR, especially in a competitive environment where other blockchains are equally pursuing scalability and AI application goals. A disciplined approach entails viewing NEAR as a high-conviction gamble on a specific yet unproven future, appropriately sized against the thesis being ahead of supporting evidence. Observing real usage data—including fees, active users, and agent activity—constitutes the essential assignment for determining if the narrative is maturing into reality.
This diligence is especially relevant within the current macroeconomic and liquidity context for altcoins, where strong narratives can still face challenging conditions. The technology and strategy are robust; however, the question of actual adoption looms considerable, with a watchful eye on it—rather than on price—being essential for determining if the bet yields rewards.
None of this is intended as investment advice; instead, it offers a framework for evaluating a distinctly ambitious AI venture in crypto, with a clear understanding of what has been validated and what remains speculative.
A coherent bet, ahead of its evidence
NEAR’s wager illustrates one of the clearest expressions of the AI-crypto thesis in the market: the belief that autonomous AI agents will engage in transactions on-chain at machine speed and volume, demanding a blockchain capable of meeting these criteria.
The June dynamic resharding upgrade serves as the lynchpin, providing automatic, instant scalability tailored to accommodate the unpredictable bursts expected from an agent economy. Surrounding this, NEAR has devised a cohesive strategy: cross-chain settlement through Intents, privacy infrastructure, and tokenomics engineered to funnel usage-derived fees toward acquiring NEAR tokens.
Led by a team deeply rooted in AI and aimed at a feasible future, this bet stands as specific, technically credible, and deserving of serious consideration, in stark contrast to the nebulous AI branding that saturates numerous projects.
However, the complication surfaces from the gap between the narrative and empirical evidence. NEAR’s price and story indicate a network on the verge of an AI-agent explosion, while its actual usage data—marked by a sharp decline in daily active users in 2026—presents a sobering contrast.
Some of the price surge has been propelled by market forces like short squeezes and the appeal of the AI narrative, rather than authentic adoption. The agent economy that NEAR anticipates has yet to take shape on a significant scale within its network, rendering the thesis credible yet unsubstantiated and ahead of its evidence.
A fair evaluation must acknowledge both aspects: NEAR has crafted impressive infrastructure for a promising future; however, that envisioned future is absent in the usage metrics, framing NEAR as a high-conviction bet on a distinct future rather than a reflection of present realities.
Whether the dynamic resharding and the Intents system will evolve into the foundational layer for a genuine agent economy—or whether the narrative will advance beyond mere adoption—poses the primary conundrum defining NEAR. The answer lies not in the token price, but in the actual emergence of those agents.
The bet has been placed, and the infrastructure is prepared; however, the economy it seeks to support has yet to materialize.
Frequently asked questions
What is NEAR betting on with the AI-agent thesis?
NEAR is betting that the future of crypto involves autonomous AI agents—software that transacts independently at machine speed—and that these agents will require a blockchain tailored to manage their unpredictable, high-volume activities. It envisions an on-chain economy where agents autonomously purchase computing resources, settle payments, and store data, resulting in transaction spikes that could overwhelm typical blockchains. NEAR aims to position itself as the scalable settlement layer for this agent economy, branding its token “the currency of agents.”
What is dynamic resharding?
Dynamic resharding is a NEAR upgrade scheduled for June 2026, part of network release 2.13, that allows the blockchain to automatically expand capacity when demand surges. NEAR utilizes sharding, which divides the network into parallel partitions, or shards, enabling simultaneous processing of transactions—akin to having multiple checkout lines. Historically, adding a shard necessitated a lengthy manual process for validator coordination and governance approval. Dynamic resharding eliminates this bottleneck: when a shard becomes full, it automatically splits into more shards, without human intervention, thereby adding real-time capacity necessary to accommodate AI-agent surges.
Why would AI agents need a special blockchain?
AI agents operate at machine speed in unpredictable bursts. A rush of agents responding to an opportunity can trigger hundreds of thousands of transactions quickly, resulting in congestion on conventional blockchains, which can spike fees and delay confirmations for all users. Therefore, a blockchain tailored for agents needs to absorb these surges instantaneously while maintaining low fees and rapid confirmations. Dynamic resharding provides this crucial automatic scaling.
How does NEAR’s token capture value from this?
NEAR’s tokenomics revolve around two pivotal changes. First, it reduced its inflation rate, consequently diminishing dilution for existing holders since the supply is now fully unlocked. More critically, NEAR activated a fee mechanism within its Intents cross-chain settlement system, whereby fees collected from settlement activities are used to repurchase NEAR tokens on the open market. This initiates a feedback loop: heightened system usage yields increased fees, which drive additional demand for NEAR tokens, effectively tying the token’s value to actual network usage instead of speculative trading.
What is the problem with NEAR’s story?
The challenge lies in the disparity between the narrative and real-world outcomes. NEAR’s price and storyline suggest a thriving future centered on AI agents, but the data signals contrasting trends: daily active users dropped significantly in 2026, from nearly three million to a mere fraction. This steep decline deviates sharply from the rising price. Analysts signal this trend as a red flag. Additionally, part of NEAR’s price rally has been influenced by factors unrelated to fundamental adoption, such as short squeezes and the prevailing momentum of the AI narrative, making the unproven thesis a concern.
Is NEAR a good investment?
The viability of NEAR as an investment hinges on one’s belief in the AI-agent thesis and NEAR’s ability to execute. The case for it sustains relevance with a plausible future, robust technological capabilities, an experienced team, and tokenomics that correlate value with usage. Conversely, caution is warranted given the unproven nature of the thesis, declining user activities, the absence of a scaled agent economy, and competitors pursuing similar objectives. A meticulous investor should focus on genuine usage data, not pricing fluctuations, as the principal measure of value. This is not investment advice.
As of June 21, 2026. Crypto markets and protocol details change rapidly; please verify current data before relying on this analysis. This article is for informational purposes only and does not constitute investment advice.
