.jpg)
NEAR has crossed $70 million in confidential TVL, and the milestone tripped a wire: the first snapshot under the protocol's Intents incentive program, which sets aside 333,333 milestone tokens for eligible users. Independent tracking puts the figure around $70.8 million as of September 17, reached through Confidential Intents, NEAR's private execution system for cross-chain transactions. For a protocol betting that privacy is becoming infrastructure rather than a niche, the number is the first hard evidence the bet has liquidity behind it.
The milestone comes with strings attached, and they matter more than the headline. Rewards stay locked until NEAR's three-day volume-weighted average price reaches at least $3.33 — an unlock mechanism that ties incentive payout directly to price, in a way that rewards momentum rather than just usage. The program also caps any single wallet at 2% of the distribution, an explicit anti-whale provision that NEAR says is designed to spread the drop across more participants.
How NEAR Confidential TVL Grew
The mechanics behind the milestone are the real product story. Confidential Intents routes transactions through a private NEAR shard, removing them from public mempool exposure and, by NEAR's account, protecting users from front-running, strategy leakage and other forms of maximal extractable value. The pitch is privacy without the computational weight of zero-knowledge systems — confidential execution for high-volume swaps across connected chains without publicly linking the activity to the party behind it. It is a direct answer to the world where an MEV bot can capture $7.8 million in a single block by watching the public mempool: if the queue is private, the predatory dex-bot economy loses its line of sight.
Alex Shevchenko, general manager of NEAR Intents, framed the milestone as demand-side proof: confidentiality is quickly becoming a core requirement for the industry, he said, and NEAR is becoming the rail to make it the new default. That is a marketing gloss on real growth, but the growth is real. NEAR's broader Intents layer — the abstraction where users specify an outcome and solvers compete to execute it — had generated more than $35.4 million in cumulative fees by June, with average daily fees above $125,000 that month, and integrations now span more than 30 blockchains, including an Aptos connection enabling one-click transfers of assets like Bitcoin, Ethereum and XRP across 20-plus networks.
The AI Bet Behind the Privacy Stack
Confidential Intents is also a load-bearing piece of NEAR's AI-agent thesis, and that context deserves explanation. The protocol's argument is that autonomous software agents handling payments, trades and treasury operations need two things the current public-chain stack handles poorly: cross-chain execution without managing assets on every network, and privacy — because a trading strategy that is visible onchain is a strategy that can be copied or extracted. Privacy tools now sit alongside NEAR's staking-based AI payment system introduced in July, which lets users lock NEAR to receive monthly compute credits across 43 models from OpenAI, Anthropic and Google, and its automatic anonymization of personally identifiable information in prompts sent to closed models.
That positioning puts NEAR in the same race as every infrastructure project arguing that machine agents will need wallets the way humans have phones. The difference is NEAR has been shipping the plumbing for a while, and the confidential layer is what turns agent transactions from a demo into something an institution could actually run — the same trust question raised by projects building verifiable records for automated decisions, answered with private rails instead of public ledgers.
The Skeptical Checklist
Three cautions belong next to the milestone. First, the $70 million TVL figure measures assets routed through the confidential system, and TVL is a flow-dependent metric — incentive programs are, by design, a way to buy it. The 333,333 milestone tokens exist precisely to grow this number, so treating the number as organic demand overstates the case. Second, the $3.33 VWAP unlock makes the incentive program a price instrument: if NEAR's price stalls, the milestone tokens sit locked indefinitely, and the program's legitimacy rests on a market variable the protocol does not control. That circularity — rewards that require price appreciation to pay out — is defensible design but should be named what it is. Third, NEAR's own governance context is unsettled: the protocol has separately proposed a 30 million token fund to address inflation concerns, a reminder that token economics remain the unresolved question underneath every TVL milestone.
The balanced read is that the privacy thesis is getting real validation — $70 million did not arrive from a whitepaper — and the incentive structure is unusually disciplined by airdrop standards, with its per-wallet cap and VWAP gate. The next check is repeatability: whether confidential TVL holds after the first distribution, or deflates once the subsidy leaves. Privacy rails earn their keep when institutions stay after the candy is gone.
