In the blockchain world, when two public blockchains that usually keep to themselves suddenly fall out, it’s often not because their roadmaps are worlds apart. On the contrary, it’s because their underlying technical approaches and business narratives have collided.
Over the past two days (October 21–23), DFINITY (ICP) founder Dominic Williams launched a 48-hour barrage of attacks on NEAR Protocol and its flagship product, NEAR Intents (intent-based trading and chain abstraction), on X. He began by questioning the technology’s substance, then escalated to accusations of “corporate espionage and poaching, and plagiarism of proprietary cryptographic algorithms,” even threatening civil and criminal lawsuits and claiming a “70%–85% chance of winning.”
Many onlookers only saw the spectacle of the two communities trading insults and even launching Meme coins to mock each other. But strip away the emotional noise, and every point of contention in this fight strikes at the heart of today’s Web3 divide between “academic-minded geeks” and “pragmatic engineers,” as well as the fault line between open-source academic ideals and commercial moats.
Here’s a breakdown of the real dynamics behind this drama across four key areas of conflict:
1. The spark: A dispute over privacy mechanisms—is it “a Band-Aid” or “gradual engineering implementation”?
The first round of clashes (October 21–22) centered on Confidential Intents, the privacy feature NEAR has been promoting recently.
Dominic’s ( #DFINITY ) criticism: Selling “permissioned sharding + TEE” as cryptographic privacy is false advertising (snake oil). Dominic shared the findings of a review of NEAR Confidential Intents conducted by one of his PhD engineers, and the attack was fierce. He argued that NEAR’s so-called confidential privacy transactions don’t actually rely on robust cryptography under the hood. Instead, they depend on a separate permissioned validator shard and a cross-chain bridge built on a TEE (Trusted Execution Environment). In simple terms, transaction privacy depends entirely on “trusting these node operators not to spy or act maliciously,” and there may even be a backdoor risk.
NEAR’s official response and the community’s counterattack: The shocking secret you spent all that time investigating was written right on the first page of our public documentation. NEAR’s response was devastating: from the outset, NEAR defined Confidential Intents as an “opt-in” privacy feature (primarily for preventing MEV front-running and managing enterprise multisig treasuries, not for mixing coins). And the design using a permissioned validator shard plus a TEE bridge was clearly spelled out in NEAR’s public documentation.
The community immediately started piling on. One developer quipped: “DFINITY’s PhD spent ages digging through the code, only to conclude what NEAR’s public docs already said. It’s like a Michelin-starred chef spending ages investigating, then announcing to the world that they’ve discovered ovens get hot. (That’s a pretty savage roast.)”
NEAR also took the opportunity to mock the “armchair theorists” who are so fixated on supposedly perfect technical ideals that they’ll never be satisfied, arguing that the essence of engineering is to get real-world use cases working first. The NEAR community even launched a Meme coin called IKP to mock ICP as an “outdated L1 trying to market itself by picking a fight.”
2. Escalation: The saga of “plagiarism, poaching, and infringement” in foundational cryptography
After the first attack was met with widespread mockery, Dominic Williams dropped a second bombshell on the evening of October 22 and into the 23rd, escalating the argument from “which technical roadmap is better” to “intellectual property theft and legal action.” His allegations centered on three points:
Allegation that the core algorithm was used without permission (Threshold ECDSA presigning): Dominic claimed that after an in-depth investigation, he found the core component powering NEAR Intents and Chain Abstraction (Chain Signatures, which let users directly control BTC and ETH accounts without managing multiple chain-specific private keys) used proprietary algorithms from a DFINITY paper published in 2022 without permission—specifically, ECDSA additive key derivation and presigning algorithms (Additive Key Derivation & Presigning). This approach lets multiple devices jointly generate addresses and sign without any one of them holding the complete private key, increasing signing speed by about 3x. Dominic emphasized that the technology was developed specifically to integrate Bitcoin and Ethereum with ICP, and that its license permitted use only on the Internet Computer (ICP) network. He also accused NEAR of recently trying to downgrade the algorithm to a slower version on GitHub to avoid the risk.
Confidential Key Derivation (CKD) overlaps with vetKeys: Dominic went on to argue that the underlying principles of NEAR’s Confidential Key Derivation (CKD), released in 2025, bear a strong resemblance to DFINITY’s proprietary vetKeys technology, launched back in 2023.
Allegation of “a former employee leaking information and poaching”: Dominic claimed, unilaterally, that NEAR not only hired a former employee whom DFINITY had fired in the summer of 2021 for similar misconduct to lead development of this cross-chain signing system, but also poached two engineers from DFINITY.
Objectively, the most intriguing legal and industry paradox here is that:
NEAR’s MPC (multi-party computation) code is released under the open-source MIT License. Most ironically, when NEAR introduced the technology in its own official documentation, it openly included a link to DFINITY’s publicly available 2022 academic paper.
This raises a fundamental question for the crypto industry, one that has yet to be tested in court: Can a blockchain foundation “privatize and monopolize” the algorithmic ideas in a cryptography paper that has already been published publicly in academia? Threshold signing is now standard across cross-chain infrastructure (Lit Protocol and others use it too). Without hard evidence of direct software code plagiarism or trade secret theft, suing based solely on “the algorithmic logic came from a paper I published” is unlikely to hold up in the open-source world. It looks more like public pressure aimed at claiming “technical legitimacy and authorship.”
3. A fight over who’s truly decentralized: “8 servers” vs. “17-node MPC + AWS dependence”
Beyond intellectual property, both sides also tried to expose the other’s decentralization shortcomings:
DFINITY’s attack: Dominic criticized NEAR’s chain abstraction signing network as not truly decentralized, saying it initially relied on “about 8 servers” to keep it running. If they were compromised or colluded, the assets held in cross-chain custody would face catastrophic risk. The ICP community also accused NEAR’s front-end application layer of relying heavily on centralized cloud services such as AWS, creating a single point of failure. (ICP, by contrast, promotes full-stack on-chain hosting, with even front-end websites running inside Canister containers.)
NEAR’s well-timed counterpunch: In a highly dramatic turn, just half an hour before Dominic posted his lengthy attack, NEAR published a technical article in a remarkably well-timed move, announcing that the number of independent operators running MPC nodes on its mainnet had increased to 17. It also said they use an 11/17 signing threshold and all run inside hardware-based Trusted Execution Environments (TEEs). This “I saw your move coming” play directly countered the “8 servers” criticism.
4. Looking beneath the surface: Why did $ICP’s Dominic lose his cool at this particular moment?
If you’ve spent enough time in the capital markets, you’ll know that when industry heavyweights enter a public feud, code and papers are just the surface. Underneath, it’s always a fight over liquidity, ecosystem position, and the power to determine valuations.
Put the two blockchains’ fundamentals as of October 2026 side by side, and you’ll immediately understand why Dominic has been so angry:
Comparison dimension (latest status as of October 2026) | DFINITY (ICP — the academic-minded geek) | NEAR Protocol (the pragmatic engineering team) | Core conflict and emotional projection
Core technology roadmap | Chain Fusion + Threshold ECDSA + vetKeys + full-stack on-chain AI | Chain Abstraction + NEAR Intents + MPC/TEE + NEAR AI | The roadmaps overlap heavily: both are working on “directly controlling assets across multiple chains without cross-chain bridges” and “decentralized AI”
Underlying technical metrics (30D) | Real-time TPS is 182x NEAR’s; transaction fees are 103x cheaper; on-chain revenue is 1.7x higher | Performance is good enough; the focus is on cross-chain intent routing (Intents) and an integrated, user-friendly front end | ICP believes its underlying technology is vastly superior, yet the market response has been lukewarm
Capital and ecosystem scale | The token price is stuck around 2.70–2.90 (down over 99% from its 2021 launch price), and the ecosystem has become marginal | Market cap is 4.3x ICP’s, staking volume is 4.5x, and developer count is 1.6x | Capital markets don’t price “the difficulty of the research paper”; they price “narrative and liquidity”
Recent commercial breakout moments | Armed with world-class cryptographic tools, but lacking a financial liquidity driver that can break into the mainstream | NEAR Intents transaction volume surged 14.6x year over year to $35.6 billion; recently launched NEAR Business, an enterprise treasury spanning 35+ chains; NEAR AI staking surpassed one million tokens; and it is even planning to connect to Nasdaq tokenized US stock liquidity | With a more agile engineering combination, NEAR has turned all the stories ICP most wants to tell into its own TVL and transaction volume
In summary: a clash between “scientists wielding a blunt sword” and “product managers who package things quickly”
Many projects that claim to have Tech DNA aren’t really tech-focused. But to be fair, both DFINITY and NEAR are teams with genuine engineering chops in the blockchain industry. The two have simply evolved in entirely different directions:
DFINITY (ICP) is a classic case of “cryptographic fundamentalism”: even if no one is paying attention, it will spend years developing the purest possible threshold signatures, full-stack decentralized containers, and vetKeys. But in a financial market where attention is extremely scarce, becoming too fixated on perfection at the foundational layer often means missing the window for ecosystem growth. When Dominic sees that the presigning algorithm his team painstakingly developed in 2022 was taken by NEAR next door and combined with TEE to build “NEAR Intents”—which then generated $35.6 billion in transaction volume within a year, while its stock price rose 5% against the market trend—it’s completely understandable that he’d feel, “I worked so hard to plant the tree, and you casually picked the fruit and got more attention than me.”
NEAR, by contrast, is the quintessential “pragmatic product manager”: it doesn’t matter who first published the algorithm in a paper—as long as it’s open source and usable, NEAR will rapidly combine “MPC + permissioned sharding + TEE” to build the cross-chain intent trading (Intents) and AI staking frameworks that institutions and retail users urgently need. Even if it initially had only 8 nodes, and even if it needed TEE as a stopgap, the priority was to get funds and transaction volume flowing, then upgrade the network to 17 nodes along the way.
Viewed through the lens of game theory, this dispute is unlikely to end with a real courtroom victory for either side (because it is extremely difficult under the law to claim an exclusive monopoly over publicly available cryptography papers). Instead, the equilibrium will likely be found in the ecosystem niches each side occupies. For DFINITY, even though its attack has earned it accusations of “marketing by picking a fight,” at least it has drawn the market’s attention back to ICP’s strengths in threshold signatures and real-time throughput. For NEAR, as long as there is no actual private-key security breach, this “pixel-level technical scrutiny” from a competitor’s founder may actually serve as a catalyst for free publicity for NEAR Intents’ $35.6 billion in transaction volume.
To get a more direct understanding of the NEAR Intents architecture at the center of this media storm, its privacy transaction logic, and the firsthand explanation from its co-founder, watch Illia Polosukhin explain NEAR Intents, privacy, and post-quantum security. In the video, NEAR’s co-founder gives the clearest official perspective on the cross-chain intent routing and privacy design that sparked this controversy.
Objectively speaking, ICP’s excessive valuation and token price at launch, along with the very heavy bag many investors were left holding, have also contributed to its prolonged slump over the years. But as an airdrop farmer, I have to say that ICP’s 2018 airdrop—where a single email address could get you a few hundred to a few thousand dollars—is something the world is unlikely to see again. #Crypto
My personal feeling about ICP is that it always seems to fall just short. You can’t say it doesn’t do marketing—it often invests heavily in major events—but the product is just difficult to use. There isn’t even a good, easy-to-use compatible wallet that makes operating on the ICP chain feel as intuitive as using EVM or Solana.
ICP has all the usual problems of a technology-focused project. Algorand, HBAR, and Conflux have them too. Algorand has a new CEO now, though who knows how things will go this time. As someone who was an early participant in $ALGO, NEAR, and ICP, my impression is that the CEO at “Er Gouzi” changes the most often, its technology is the most hardcore, and it’s the worst at marketing. ICP is somewhere in the middle. NEAR has connections to Nvidia’s Jensen Huang and Google, so it has an advantage.
$NEAR ’s overall strategic plan is clearly pragmatic, and at this critical juncture it can also integrate more quickly with #AI and #Agent. That’s extremely valuable.
This content is for informational and research reference only and does not constitute any investment, trading, or financial advice. Analysis & Polish & Formatting: UniPublisher