StarkWare launches the S-Two fitting room on Starknet, expanding privacy and scalability.
- StarkWare implements S-two for increased speed on Starknet.
- Provider allows private proof on consumer devices.
- Update accelerates and decentralizes DeFi infrastructure at layer 2.
StarkWare has announced the implementation of its next-generation zero-knowledge prover, S-two, on the Starknet core network, focusing on reducing costs, latency, and increasing decentralization. According to the company, S-two represents "the fastest and most privacy-protected production proving system in the world."
The upgrade replaces the previous network verification components and now generates validity proofs for each block, reducing verification time and validation costs. Simultaneously, network throughput and scalability improve without compromising reliability. The system also allows independent operators to collaborate with computational resources, increasing the network's resistance to censorship.
In Layer 2 environments, such as Starknet, provers play an essential role: they generate cryptographic proofs that attest to the correctness of off-chain transactions before settlement at the base layer. This architecture allows the network to scale without relying on heavy on-chain computations, reducing fees and improving efficiency.
The S-two, short for "STARK Two," was designed to be up to ten times faster than previous versions—in performance tests, it outperformed alternatives by a significant margin. One of its key differentiators is its ability to run on consumer hardware, such as smartphones and laptops, opening the way for privacy applications and everyday use.
“We created S-two with a clear objective: to improve the user experience on Starknet.”
"said Eli Ben-Sasson, co-founder and CEO of StarkWare."
“When tests are so quick and economical, new categories of applications become viable, and decentralization ceases to be an aspiration and becomes an inspiration…”
In addition to boosting Starknet's scalability within the Ethereum ecosystem, StarkWare highlights that S-two will pave the way for new frontiers: on-chain privacy, anonymous identity, verifiable AI inference, and games with zk-based proof security. The launch is presented as a cornerstone of what the company defines as a "financialization layer" that also includes settlement in both Bitcoin and Ethereum through a single, privacy-ready execution layer.
With the S-two present, Starknet reinforces its position in the crypto-asset and DeFi space as a layer 2 infrastructure capable of meeting both high-performance and privacy demands, while preparing for new use cases in common hardware and decentralized operations.
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
AI bubble alarm rings across Wall Street! Oaktree Capital's Marks disagrees: The market hasn't reached its most dangerous moment yet, and investors are not completely oblivious to risks.
Although investors are displaying typical FOMO (fear of missing out) sentiment regarding further gains in tech stocks, the market continues to draw comparisons between the current situation and past bubbles, indicating that investors are not unaware of the risks.

Barclays: Dual restrictions of physical and capital, data center growth will slow significantly in 2027!
Barclays' latest report reveals that the global data center construction boom is approaching a turning point—the year-on-year growth rate of new capacity will plummet from over 100% in 2026 to about 35% in 2027, and stabilize by 2028. Facing both physical constraints and tightened financing, hyperscale cloud providers will absorb more than 60% of new capacity, while the risk of project suspension for small and medium developers will sharply increase.

Bitcoin and Solana test key support as NEAR and HYPE face selling pressure
AI Capital Flow Chart: Nvidia Takes 690 Billion, Remaining 499 Companies Share 340 Billion
Morgan Stanley has released its first comprehensive overview of AI capital expenditure: In 2027, among $1.58 trillion in spending by hyperscale cloud service providers, Nvidia alone will take 67%, while the remaining 499 S&P component companies will collectively receive only $339 billion. A more hidden risk lies in over $1.1 trillion of "off-balance-sheet leases," which are essentially disguised debt. 2027 is predicted to have the largest free cash flow shortfall, and the risks of both volume and price declines are being severely underestimated by the market.

