From Aztec to Zcash: Privacy Evolving from 'Gray Industry Tool' to 'Institutional Necessity'

比推Dipublikasikan tanggal 2025-12-26Terakhir diperbarui pada 2025-12-26

Abstrak

From Aztec to Zcash: The Rise of Pragmatic Privacy in Blockchain In 2025, blockchain privacy evolved from a niche concern to a mainstream priority, driven by the concept of "pragmatic privacy" that balances individual anonymity with regulatory compliance. This shift is marked by the strong performance of privacy-focused assets like Zcash and new institutional initiatives. Several new privacy-centric blockchains launched or advanced significantly. Aztec Network's Ignition L2 mainnet went live, raising over 6,100 ETH. Nillion launched its "blind computer" mainnet for encrypted data computation. Cosmos-based Namada introduced a "composable privacy" L1, while Miden spun out from Polygon to build its own privacy chain. Umbra raised $154.9 million in an ICO to build on Solana. Horzen transitioned to a Base L3 privacy solution. Institutions are actively embracing privacy. Coinbase hired a team from Iron Fish to develop privacy primitives for Base. Circle is testing a privacy-preserving wrapped USDC (USDCx) on Aleo. The Canton Network and EY's Nightfall L2 are focusing on confidential enterprise solutions. The Ethereum Foundation formed a dedicated "Privacy Cluster" and an "Institutional Privacy Working Group." Vitalik Buterin also launched Kohaku, an open-source wallet framework for compliant privacy. At the application layer, tools like 0xbow's Privacy Pools (based on Vitalik's research) and Railgun (using "proof of innocence") allow users to anonymize transactions without aidi...

Source: The Block

Original Title: From Aztec to Zcash: The year 'pragmatic privacy' took root

Compiled and Edited by: BitpushNews


Since Satoshi Nakamoto wrote the Bitcoin whitepaper, the privacy limitations of blockchain have been evident. The cryptocurrency pioneer pointed out at the time: despite using anonymous mechanisms, Bitcoin addresses could still be traced back to real-world identities.

For many years since, this "pseudo-anonymous" characteristic has been tacitly accepted as "acceptable enough," causing most blockchain privacy projects to remain on the margins for a long time.

However, this year has seen a renewed interest in blockchain privacy, with Zcash becoming one of the best-performing assets of the year, and the Ethereum Foundation launching multiple end-to-end encryption initiatives. The concept of "Pragmatic Privacy" has also begun to take root, aiming to balance individual privacy with compliance considerations.

Emerging Privacy Chains

2025 has witnessed the launch of several new privacy-focused blockchains at various stages of development.

Perhaps the most notable is the Aztec Network's Ignition chain going live on mainnet in November. This is not only because it is the "first fully decentralized L2 on Ethereum," with multiple innovations in consensus and ZK technology, but also because it has overcome numerous obstacles since the project's inception in 2017.

Earlier this month, Ignition also raised 19,476 ETH (approximately $61 million) from 16,741 participants using a new "Continuous Clearing Auction" mechanism co-developed with Uniswap Labs.

Nillion is another blockchain project that launched its mainnet in 2025. This so-called "blind computer" aims to perform computations on encrypted data and has already integrated with multiple networks, including Layer 1's Near and Layer 2's Arbitrum, to enhance privacy at the application layer.

Namada, built on Cosmos, also launched its Layer 1 mainnet in June, focusing on "composable privacy" and supporting multiple blockchain ecosystems—including Bitcoin L2s (like Lombard and Babylon), Ethereum, and Solana—via cross-chain bridges.

Miden, which has been in development since at least 2022, is preparing to launch its mainnet next year. Even so, the project took a huge step towards independence this year: spinning off from Polygon and raising $25 million to build the privacy-focused blockchain Edge.

Similarly, the privacy protocol Umbra, powered by Arcium, conducted its Initial Coin Offering (ICO) on MetaDAO in October, raising $154.9 million. Umbra plans to launch concurrently with Arcium's mainnet Alpha, becoming one of the first privacy protocols built on Solana utilizing its infrastructure.

Horizen, one of the oldest crypto privacy projects, also underwent a major transformation this year: it shut down its proprietary Layer 1 and L2 incubator and relaunched as a Layer 3 privacy solution on Base, the network incubated by Coinbase.

Grayscale, which manages the ZEN token fund, also filed a registration statement with the U.S. Securities and Exchange Commission (SEC) to convert its Grayscale Zcash Trust into the first ZEC ETF. Zcash, which launched the Zebra 3.1 upgrade, has been one of the best-performing tokens in the second half of the year.

Institutional Privacy Initiatives

Coinbase signaled multiple times this year that privacy is not optional for Base. Earlier this year, the company announced it had hired a development team from Iron Fish (a long-standing PoW privacy project) to provide "privacy-preserving primitives" for Base.

Circle, the largest U.S. stablecoin issuer, also made waves by announcing it was testing a privacy-preserving wrapped USDC called USDCx on the Aleo testnet.

USDCx appears to focus on "configurable compliance," aiming to support enterprise-level use cases such as payroll management and e-commerce.

Meanwhile, the Canton Network, backed by Goldman Sachs and BNY Mellon, and Nightfall, an Ethereum L2 incubated by EY, focus on enterprise blockchain solutions that require extreme confidentiality. Nightfall has already announced support for two blockchains: the real-world asset (RWA) network Plume and CELO.

The Ethereum Foundation initiated multiple privacy programs this year, signaling that the topic has moved from a niche research area to a core concern for Ethereum developers. In October, the Foundation formed the so-called "Privacy Cluster," a dedicated team of dozens of engineers aimed at consolidating its privacy efforts, including a roadmap for achieving native end-to-end encryption. The Foundation also launched the "Institutional Privacy Working Group" to guide institutions and enterprises into Ethereum.

Ethereum's Vitalik Buterin also introduced Kohaku, an open-source centralized wallet framework for privacy, aimed at fostering "default and compliant" privacy features.

Pragmatic Privacy

The concept of "pragmatic privacy" is perhaps most fully realized at the application layer. This year, 0xbow launched Privacy Pools, a tool for ordinary blockchain users to erase transaction records without running into legal trouble.

Privacy Pools are based on research published by Buterin and several other senior cryptographers on "association lists," designed to prevent bad actors from profiting from crypto mixers. The 0xbow team recently raised a $3.5 million seed round, expanded support for Sky's USDS stablecoin, and provided a method for Tornado Cash users to remain anonymous "without helping hackers." It is also part of the Ethereum Foundation's Kahaku software framework.

Similarly, Railgun is listed in Kahaku's GitHub repository as a means of shielding funds. Railgun achieves blockchain privacy through a "proof of innocence" system, where users generate ZK proofs showing their funds/transactions do not belong to a preset list of flagged/malicious addresses.

Finally, it's worth noting that as a standard-bearer for crypto privacy, Zcash's anonymous pool supply has grown to nearly 25%, marking a steady increase in network privacy adoption. Zcash provides users with functional privacy and the ability to achieve compliance through selective disclosure of information.


Twitter:https://twitter.com/BitpushNewsCN

Bitpush TG Discussion Group:https://t.me/BitPushCommunity

Bitpush TG Subscription: https://t.me/bitpush

Original link:https://www.bitpush.news/articles/7598558

Pertanyaan Terkait

QWhat is the main concept that has gained traction in blockchain privacy in 2025, as discussed in the article?

AThe main concept is 'Pragmatic Privacy', which aims to balance individual privacy with compliance considerations.

QWhich privacy-focused blockchain launched its mainnet in November 2025 and is noted as 'Ethereum's first fully decentralized L2'?

AAztec Network's Ignition chain launched its mainnet in November 2025 and is described as Ethereum's first fully decentralized L2.

QWhat significant step did the Ethereum Foundation take in October 2025 regarding privacy?

AIn October 2025, the Ethereum Foundation formed a 'Privacy Cluster', a dedicated team of dozens of engineers to consolidate its privacy efforts, including a roadmap for native end-to-end encryption.

QWhat is the purpose of Privacy Pools, as mentioned in the article?

APrivacy Pools is a tool for ordinary blockchain users to erase transaction records without getting into legal trouble, based on research about 'association lists' to prevent bad actors from benefiting from crypto mixers.

QWhich stablecoin issuer is testing a privacy-preserving wrapped version of USDC called USDCx on the Aleo testnet?

ACircle, the largest U.S. stablecoin issuer, is testing a privacy-preserving wrapped version of USDC called USDCx on the Aleo testnet.

Bacaan Terkait

Model Generatif Bisa Dilatih End-to-End? Intinya Hanya Sebuah For Loop

Model generatif seperti model difusi atau autoregresif yang dominan saat ini biasanya tidak melatih seluruh proses generasi secara end-to-end. Mereka dilatih hanya untuk memprediksi satu "langkah kecil" dalam proses, namun saat inferensi, langkah ini harus diulang ratusan atau ribuan kali. Ketidaksesuaian ini menyebabkan masalah "exposure bias," di mana kesalahan menumpuk seiring waktu. Makalah terbaru dari UIUC dan Harvard memperkenalkan paradigma baru bernama **Explorative Modeling (XM)**. Inti dari metode ini sangat sederhana: alih-alih menghasilkan satu sampel, model menghasilkan **K kandidat** dalam setiap langkah pelatihan. Kemudian, hanya kandidat yang paling dekat dengan data target yang dipilih untuk menghitung gradien dan memperbarui model. Mengapa ini berhasil? Dalam tugas generatif, satu input bisa memiliki banyak output yang valid (mode). Fungsi kerugian rekonstruksi tradisional cenderung menyebabkan "mode blurring," di mana model hanya mempelajari rata-rata dari semua mode, menghasilkan output yang tidak realistis. Dengan menghasilkan beberapa kandidat, model dapat "mengeksplorasi" dan menangkap beberapa mode yang berbeda sekaligus, sehingga menghindari rata-rata yang kabur. Penelitian ini menunjukkan bahwa **"eksplorasi" (K) bertindak sebagai sumbu penskalaan ketiga** yang efektif, selain parameter dan data. Eksperimen pada gambar, video, dan bahasa menunjukkan peningkatan kinerja yang monoton dengan peningkatan K, dengan keuntungan yang lebih besar pada skala yang lebih besar. XM dapat meningkatkan efisiensi FLOP hingga 4.1x dan efisiensi sampel hingga 6.2x. Yang lebih menarik, ketika diterapkan secara ekstrem, XM memungkinkan **pelatihan generatif yang benar-benar end-to-end**. Dalam tugas kontrol robot, "Explorative Policy" berbasis XM mencapai kinerja yang setara dengan "Diffusion Policy" yang membutuhkan 100 langkah inferensi, sementara hanya membutuhkan satu langkah inferensi maju. Ini membuktikan bahwa kompleksitas untuk menangani banyak mode dapat dipindahkan dari inferensi ke pelatihan. Meskipun ide "best-of-K" bukan hal baru, kontribusi utama makalah ini adalah penjelasan teoritis bahwa mekanisme ini secara langsung meningkatkan **ekspresivitas generatif** model. Dengan demikian, ini menawarkan cara baru yang ampuh untuk mengatasi tantangan mendasar dalam pemodelan generatif.

marsbit5m yang lalu

Model Generatif Bisa Dilatih End-to-End? Intinya Hanya Sebuah For Loop

marsbit5m yang lalu

OpenAI Tidak Lagi Mengandalkan Model Termahal untuk Menghasilkan Uang

OpenAI tidak lagi mengandalkan model termahal untuk menghasilkan uang. Pada 30 Juli, perusahaan mengumumkan penyesuaian harga, menurunkan harga model GPT-5.6 Luna sebesar 80% dan Terra sebesar 20%. Yang lebih penting dari angka penurunan harga adalah pesan intinya: untuk banyak tugas, model terkuat (seperti Sol) tidak selalu diperlukan. OpenAI secara eksplisit merekomendasikan strategi di mana model mahal (Sol) menangani perencanaan dan analisis kompleks, sementara model yang lebih terjangkau (Luna) mengeksekusi tugas, menulis kode, dan menjalankan pengujian. Pergeseran ini mencerminkan perubahan industri AI dari hanya mengejar kecerdasan tertinggi menuju optimisasi biaya dan efisiensi untuk penggunaan skala besar. Anthropic juga mengikuti pola serupa dengan meluncurkan Claude Opus 5 dengan harga setengah dari model andalannya, Fable 5. Kedua raksasa Silicon Valley ini menunjukkan bahwa model flagship kini berfungsi lebih untuk nilai merek dan pembuktian teknologi, sementara model "kendaraan massal" yang lebih murah (seperti Luna, Terra, Opus) yang akan mendorong adopsi komersial luas dan menghasilkan pendapatan utama. OpenAI mengungkapkan bahwa penurunan biaya ini sebagian didorong oleh model AI (Sol) yang mengoptimalkan kode produksi dan alur kerjanya sendiri, menciptakan siklus peningkatan efisiensi yang dapat mempercepat penurunan biaya di masa depan. Intinya, kompetisi AI bergeser dari "siapa yang paling pintar" menjadi "mana yang paling bernilai". Perusahaan tidak lagi membeli satu model tunggal, tetapi merancang sistem dengan beberapa model yang cocok untuk tugas yang berbeda—mirip dengan arsitektur komputasi awan. Tujuan akhir OpenAI adalah membangun ekosistem yang mengunci alur kerja pengembang melalui volume panggilan API yang sangat besar dan biaya yang sangat rendah, sehingga menciptakan daya tarik yang kuat. Ketika AI menjadi murah dan tersebar luas seperti listrik, tertanam dalam setiap proses tanpa diperdebatkan, era sesungguhnya baru dimulai.

marsbit1j yang lalu

OpenAI Tidak Lagi Mengandalkan Model Termahal untuk Menghasilkan Uang

marsbit1j yang lalu

Trading

Spot
活动图片