Decentralizing the Next Layer of Ethereum Infrastructure with Anti-Slashing & ZK-Readiness

Decentralizing the Next Layer of Ethereum Infrastructure with Anti-Slashing & ZK-Readiness

At the Scaling Summit Singapore, a pivotal conversation unfolded on the Ethereum Stage, where builders, researchers, and visionaries gathered to confront one of the ecosystem’s most pressing dilemmas: How do we scale Ethereum without sacrificing its foundational ethos of decentralization? Moderated by Luca Donno, a researcher at L2Beat, the panel featuring Amir (Puffer Finance), Mike Massari (Redstone), Ian Wallis (Linea), and Anndy Lian (Intergovernmental Blockchain Advisor) delved into the tension between idealism and pragmatism in blockchain infrastructure.

The Centralization Conundrum

The discussion opened with a stark reality: while decentralization remains Ethereum’s “biggest asset,” market forces often incentivize centralization for speed and user experience. As Amir of Puffer Finance noted, “If you look at where biggest asset holders are now parking their assets… they’re trusting Ethereum for a reason.” He pointed to USDT and USDC 45% and nearly 100% of their supplies, respectively, reside on Ethereum precisely because of its trustless nature.

The path to mass adoption is rarely pure. Luca framed the dilemma: “We were very much in a situation in which decentralization was the most important thing… Now it’s not anymore. That is not the focus of institutions.” This shift demands a recalibration. Anndy Lian, speaking from a macroeconomic lens, admitted bluntly: “Most users, including VCs like myself, you know, we don’t really care [about decentralization]… we want to make money.” His candid remark underscored a broader truth user incentives today prioritize yield and UX over ideological purity.

But the panelists agreed: decentralization must remain the north star, even if the journey begins with centralized stepping stones. “It is okay to start slightly more centralized,” Amir argued, “but having decentralization on the roadmap as the main goal is the only way we can scale the entire blockchain to its full capacity.”

Anti-Slashing: Guardrails for a Risky Landscape

A key innovation discussed was anti-slashing a critical safeguard in the era of liquid staking tokens (LSTs). With LSTs now dominating Ethereum’s staking landscape, systemic risk looms large. As Luca observed, many protocols hold more LSTs than native ETH, creating concentration points that threaten network security.

Amir explained how Puffer Finance addresses this: “We didn’t stop at permissionless restaking. We launched bonded validators operators must stake their own capital. If slashing occurs, it’s their money on the line.” This “skin in the game” model, combined with hardware-based anti-slashing modules (like trusted execution environments, or TEEs), prevents malicious or accidental validator misbehavior. “These modules act like a Ledger wallet,” Amir said, “but even more restricted you can only sign permitted transactions.”

Mike Massari echoed the sentiment: “The moment you detach risk from the person managing the capital, you create systemic risk.” Anti-slashing, therefore, isn’t just technical it’s economic alignment.

Ian Wallis added context from Linea’s perspective, noting their plan for a “native yield” bridge that stakes ETH directly, reducing reliance on dominant LST providers like Lido. “We’re consulting closely with the Ethereum Foundation,” he said, emphasizing collaboration over competition in securing the ecosystem.

ZK: Promise, Peril, and Patience

The conversation then turned to zero-knowledge (ZK) technology the cornerstone of Ethereum’s scaling roadmap. While optimistic about ZK’s potential, the panelists acknowledged its immaturity. “ZK is still experimental,” Luca warned, citing recent bugs in foundational libraries like Circom and Halo2. “A multi-billion-dollar bug on Ethereum L1 could shatter trust in the entire paradigm.”

Amir, however, offered a solution in progress: “We’re researching 2FA for ZK running a full Ethereum client inside a TEE alongside the ZK prover. If outputs mismatch, you halt the transaction.” This dual-verification approach could catch bugs before they cascade.

Ian, whose team at Linea operates a ZK-EVM rollup, remained bullish: “Compare where we were five years ago to now we’re light years ahead. ZK improvements are coming quarterly.” He pointed to Swift’s recent partnership with Linea as validation: “If the kings of centralized finance see potential here, that’s an endorsement.”

Anndy Lian urged patience: “Give the technology time. The big boys are coming. Adoption will follow.”

Toward a Redistributed Future

Ultimately, the panel converged on a shared vision: Ethereum must evolve progressively. As Luca summarized, “We shouldn’t decentralize for decentralization’s sake but where user funds are at stake, decentralization equals security, and security equals good UX.”

The road ahead involves balancing short-term pragmatism with long-term principles. Whether through anti-slashing economics, ZK verifiability, or middleware that enforces decentralization standards, the goal remains clear: build infrastructure that can onboard trillions not just billions without compromising Ethereum’s soul.

As Amir put it: “If we want to bring repo markets or supply chains onchain, it has to be fully decentralized and secure. Hyperliquid won’t cut it for JP Morgan.”

In that spirit, the Scaling Summit didn’t just showcase technology it reaffirmed a covenant: scale with integrity, or don’t scale at all.

 

Anndy Lian is an early blockchain adopter and experienced serial entrepreneur who is known for his work in the government sector. He is a best selling book author- “NFT: From Zero to Hero” and “Blockchain Revolution 2030”.

Currently, he is appointed as the Chief Digital Advisor at Mongolia Productivity Organization, championing national digitization. Prior to his current appointments, he was the Chairman of BigONE Exchange, a global top 30 ranked crypto spot exchange and was also the Advisory Board Member for Hyundai DAC, the blockchain arm of South Korea’s largest car manufacturer Hyundai Motor Group. Lian played a pivotal role as the Blockchain Advisor for Asian Productivity Organisation (APO), an intergovernmental organization committed to improving productivity in the Asia-Pacific region.

An avid supporter of incubating start-ups, Anndy has also been a private investor for the past eight years. With a growth investment mindset, Anndy strategically demonstrates this in the companies he chooses to be involved with. He believes that what he is doing through blockchain technology currently will revolutionise and redefine traditional businesses. He also believes that the blockchain industry has to be “redecentralised”.

j j j

The Great Infra Wars: How Web3 is Forging the Future of Decentralized AI | Taipei Blockchain Week 2025

The Great Infra Wars: How Web3 is Forging the Future of Decentralized AI | Taipei Blockchain Week 2025

Taipei, Taiwan – Sept 2025 – As artificial intelligence reshapes the digital landscape, a critical battle is unfolding beneath the surface: the fight to build the infrastructure capable of hosting truly decentralized AI. At Taipei Blockchain Week 2025, the panel “Infra Wars: The Battle to Host the AI-powered Web” cut through the hype, revealing the profound technical and philosophical challenges at the intersection of Web3 and AI. Moderated by Lee Ting Ting, Founder of FansNetwork, the session brought together infrastructure pioneers to dissect how blockchain can solve AI’s most pressing limitations, from computational bottlenecks to data sovereignty crises.

The Speed vs. Decentralization Dilemma: Rethinking Consensus

The panel opened with a fundamental tension: AI demands blistering speed, while blockchain prioritizes decentralization, often at the cost of performance. “We all know AI models have immense complexity, and users care about speed,” noted moderator Lee Ting Ting, framing the core conflict. “How are emerging consensus mechanisms being redesigned to handle AI’s computational demands?”

Jiahao Sun, CEO of Flock.io and a former financial infrastructure lead, argued that traditional blockchain architectures are fundamentally mismatched for AI workloads. “The public chain design predates the AI boom,” he explained. “Even if on-chain transaction speed is fast, a single consensus layer cannot solve the demands of AI.” Sun’s solution lies in modular consensus: “We’re using a multiple and modular consensus mechanism. We built single processors for decentralized storage and computing, but we align all different modules, data service, cloud service, and computation on top of a PoS system. This creates unlimited transaction possibilities and aligns computing with storage.”

Anthurine Xiang of Quarkchain added nuance, distinguishing between monolithic (e.g., Solana) and modular (e.g., Ethereum) chains: “For modular ecosystems, we need a shared data availability (DA) layer. Solutions like Celestia or EigenDA help store data on-chain forever, making it traceable and preventing losses like the infamous NFT storage failures.” Her point was stark: “When centralized storage fails, like when a team stops paying for AWS, your NFTs become broken links. For AI, this is unacceptable.”

JT Song of 0G Labs (ZG) took this further, announcing their new IFT standard (likely “Immutable File Token”): “For AI agents, all data must be stored on our decentralized service and trace the entire training process. This makes data verifiable and tradeable on-chain, a radical shift from traditional ERC-721.” Crucially, Song revealed ZG’s collaboration with China Mobile: “We ran decentralized training for a 100-billion-parameter model faster than centralized alternatives. Decentralized computing isn’t slower, it’s a different paradigm.”

Data Sovereignty: The Privacy Imperative

The conversation pivoted to AI’s data crisis: Big Tech’s monopolization of user data for training models. “How can infrastructure enable true user ownership while allowing decentralized training?” asked Lee.

Jiahao Sun spotlighted federated learning – a Google-originated technique now supercharged by blockchain. “Your phone predicts your typing locally; raw data never leaves your device. But Google controls the aggregation – it’s still centralized. Blockchain changes this: none of the users’ raw data is ever submitted. Instead, we submit model gradients – changes to the AI itself – which merge into a larger model. Everything is transparent on-chain.” He emphasized the breakthrough: “You don’t have to trust a third party; you see the transactions.”

JT Song reinforced this with ZG’s vision: “We’re building full-chain data services. If an AI project uses our IFT standard, all training data is stored in a decentralized manner. Even if the operation team disappears, the AI agent and its data remain self-sovereign and verifiable.” This tackles the “black box” problem of open-source AI: “Models claim transparency, but the data and process remain hidden. Blockchain forces process transparency.”

Anndy Lian, Intergovernmental Blockchain Advisor, injected pragmatism: “Full decentralization remains a big challenge. Security must be managed effectively, no hacks, no losses. But I’ve discussed zero-data AI architecture with Southeast Asian governments. Blockchain can enforce rules and enable fair audits, creating a win-win for AI and Web3.”

The Killer App: Why Decentralized AI Isn’t Optional

The panel’s most heated debate centered on the “killer app” for decentralized AI: Why bother with Web3 when centralized AI works?

Jiahao Sun targeted enterprise pain points: “Privacy isn’t just ‘nice to have’, it’s necessary in banking, healthcare, and public sectors. But mass adoption needs retail applications. Imagine a virtual companion where conversations are secured on-chain. You know no one, not even the platform, can access your private chats. That’s a healing application blockchain enables.”

Anthurine Xiang pushed for Web3’s evolution beyond finance: “Ethereum aimed to be a ‘world computer,’ but most apps are still token-trading. We need diversified use cases: AI agents, decentralized content platforms. Our ‘supercomputer’ infrastructure must enable non-financial apps with mass appeal, faster speeds, more capacity, lower costs.”

JT Song unveiled ZG’s “Air Wars” AI agent marketplace (boasting 2.3 million testnet users): “Agents can evolve, be verified, and classified. This isn’t just about functionality, it’s about ownership. Users control their AI’s data and evolution.”

But Anndy Lian delivered the most provocative insight: “The best way to onboard people to AI + Web3? Teach them how to make money. AI agents that help users make smart trades or generate income will drive adoption faster than ideology. And let’s be honest: today’s ‘Web3’ isn’t truly decentralized. We need Web4, a more decentralized, less controlled, AI-driven future.”

The Road Ahead: Beyond the Hype

As the session concluded, a clear consensus emerged: The “infra wars” aren’t about which chain wins, but how Web3’s core innovations – decentralization, transparency, and user sovereignty – can solve AI’s existential flaws. Federated learning plus blockchain enables private AI training; modular data layers prevent catastrophic data loss; and new consensus models unlock scalable compute.

The panelists acknowledged the journey is nascent. “Papa, this will be a slow process,” admitted JT Song. Anndy Lian tempered expectations: “From a productivity standpoint, putting everything on-chain remains challenging. But give us time.”

The most profound takeaway? Decentralized AI isn’t a niche experiment, it’s the only path to an AI future where users own their data, models are transparent, and infrastructure serves people, not platforms. As Jiahao Sun succinctly stated: “We’re not just building faster chains. We’re rebuilding the entire operating system for decentralized AI.”

In the battle for AI’s soul, Taipei Blockchain Week 2025 made one thing clear: Web3’s infrastructure warriors aren’t just participants in the AI revolution, they’re building its foundation. The “infra wars” have just begun, but the stakes, a truly user-owned digital future, couldn’t be higher. As Lee Ting Ting closed the session: “This isn’t about technology alone. It’s about who controls the future.” With 2.3 million testnet users already engaging with decentralized AI agents, that future may arrive sooner than we think.

Anndy Lian is an early blockchain adopter and experienced serial entrepreneur who is known for his work in the government sector. He is a best selling book author- “NFT: From Zero to Hero” and “Blockchain Revolution 2030”.

Currently, he is appointed as the Chief Digital Advisor at Mongolia Productivity Organization, championing national digitization. Prior to his current appointments, he was the Chairman of BigONE Exchange, a global top 30 ranked crypto spot exchange and was also the Advisory Board Member for Hyundai DAC, the blockchain arm of South Korea’s largest car manufacturer Hyundai Motor Group. Lian played a pivotal role as the Blockchain Advisor for Asian Productivity Organisation (APO), an intergovernmental organization committed to improving productivity in the Asia-Pacific region.

An avid supporter of incubating start-ups, Anndy has also been a private investor for the past eight years. With a growth investment mindset, Anndy strategically demonstrates this in the companies he chooses to be involved with. He believes that what he is doing through blockchain technology currently will revolutionise and redefine traditional businesses. He also believes that the blockchain industry has to be “redecentralised”.

j j j

Global Game Theory: The Response to America’s Changing Bitcoin Policy- Bitcoin Conference Asia, Hong Kong

Global Game Theory: The Response to America’s Changing Bitcoin Policy- Bitcoin Conference Asia, Hong Kong

Imagine the United States positioning itself as a dominant force in Bitcoin: What ripple effects would that create worldwide? Join host Grant McCarty from the Bitcoin Policy Institute as he moderates a discussion with experts Jeremy Tan, Nenter Chow, Anndy Lian, and Bilal Bin Saqib. Covering topics like Pakistan’s Bitcoin holdings, Singapore’s adaptation tactics, and Bitcoin’s influence on widespread adoption, this panel delves into Bitcoin’s evolving international strategic dynamics.

00:00 Intro & Panelist Introductions
02:20 Shifts in US and Global Bitcoin Policy
05:20 Reactions from Around the World
08:08 Impact on Private Sector & Financial Markets
12:40 Building Strategic Bitcoin Reserves
14:39 Challenges and Opportunities for National Bitcoin Strategies
17:36 Foundations for Bitcoin Economies
20:00 Importance of Financial Literacy and Education
23:03 Educating Lawmakers and the Public
25:01 Local and Global Pressures on Policy Formation
27:44 International Cooperation & Policy Needs
29:37 Calls for Standardized Policy and Global Frameworks
31:02 Final Thoughts & Recommendations
35:01 Panel Wrap Up & Closing

 

#BitcoinAsia2025 #Bitcoin #BTC #BitcoinPolicy #BitcoinReserve #BitcoinFuture #GlobalBitcoin #BitcoinConference #BitcoinDiplomacy #BitcoinEducation

https://asia.b.tc/speaker/anndy-lian

 

Anndy Lian is an early blockchain adopter and experienced serial entrepreneur who is known for his work in the government sector. He is a best selling book author- “NFT: From Zero to Hero” and “Blockchain Revolution 2030”.

Currently, he is appointed as the Chief Digital Advisor at Mongolia Productivity Organization, championing national digitization. Prior to his current appointments, he was the Chairman of BigONE Exchange, a global top 30 ranked crypto spot exchange and was also the Advisory Board Member for Hyundai DAC, the blockchain arm of South Korea’s largest car manufacturer Hyundai Motor Group. Lian played a pivotal role as the Blockchain Advisor for Asian Productivity Organisation (APO), an intergovernmental organization committed to improving productivity in the Asia-Pacific region.

An avid supporter of incubating start-ups, Anndy has also been a private investor for the past eight years. With a growth investment mindset, Anndy strategically demonstrates this in the companies he chooses to be involved with. He believes that what he is doing through blockchain technology currently will revolutionise and redefine traditional businesses. He also believes that the blockchain industry has to be “redecentralised”.

j j j