Web3 Infrastructure Enters a New Era: How DEX Liquidity Optimization and AI Privacy Computing Are Reshaping On-Chain Applications

On July 24, 2026, the crypto market as a whole remained in a consolidation phase. Yet beneath the surface price swings, the Web3 infrastructure layer is undergoing a profound dual evolution: decentralized exchanges (DEXs) are entering a phase of refined liquidity competition, while privacy computing technologies are moving from academic labs into on-chain production environments.

Though these two evolutionary paths may seem distinct, they both point in the same direction: the Web3 application layer is gaining a more robust foundational base. DEX liquidity optimization addresses trading efficiency and capital utilization, while privacy computing resolves the tension between on-chain data sovereignty and regulatory compliance. When these two trends converge, on-chain applications will deliver trading experiences approaching those of centralized finance (CeFi), while offering privacy protections that traditional finance simply cannot provide.

The DEX Liquidity War Enters a New Phase

In 2026, the DEX sector is showing clear signs of market concentration and structural differentiation. Uniswap remains the market leader with $3.02 billion in total value locked (TVL) and roughly $36 billion in monthly trading volume. However, this scale advantage hasn’t led Uniswap to rest on its laurels. On July 10, the Uniswap community proposed cutting liquidity provider (LP) incentives by 33% in the V4 upgrade, shifting the focus toward improving trade execution efficiency and increasing trading volume.

This proposal marks a fundamental shift in DEX competition strategy. In the V3 era, high LP incentives were the main way to attract liquidity. In the V4 era, Uniswap aims to drive trading volume growth by lowering trading costs, tightening spreads, and boosting capital efficiency—offsetting the reduction in LP rewards. The risk is clear: liquidity providers can easily move their capital to competing protocols that offer higher yields. If trading volume growth fails to make up for the cut in LP incentives, Uniswap could face a liquidity exodus.

At the same time, a Dune Analytics study has revealed efficiency bottlenecks in the concentrated liquidity model. The study analyzed the top 200 liquidity positions on Uniswap V3, V4, PancakeSwap V3, and Aerodrome Slipstream across seven blockchains from January to June 2026. It found that out of $1.84 billion in weekly liquidity, about 29.4% to 30.5% sat outside the active trading range and generated no fee income. This translates to an estimated $150 million in missed annual fee revenue. Even though Uniswap V4 introduced "hooks," which allow pools to implement custom logic and deploy idle liquidity to external protocols like Aave or Morpho for yield, very few pools are currently using this feature.

DODO’s Differentiated Path: Capital Efficiency First

In a market dominated by Uniswap, DODO has chosen a distinct technical path. DODO’s core strength lies in its proprietary Proactive Market Maker (PMM) algorithm. Unlike the constant product automated market maker (AMM) model (x × y = k) adopted by early Uniswap, the PMM algorithm leverages price oracles to obtain fair market prices and concentrates liquidity near the market price. This significantly reduces slippage and boosts capital efficiency.

According to Token Terminal, DODO leads all DEXs in the "trading volume/TVL" capital efficiency metric—meaning DODO generates more trading output with less liquidity than most competitors. However, this advantage in capital efficiency has not translated into TVL and trading volume growth. As of July 24, DODO’s token price stood at $0.01906, down 12.73% over 24 hours and 20.80% over seven days, with a market cap of about $19.61 million. Its TVL was around $12.9 million, and weekly trading volume dropped 56% over the past seven days.

DODO is working to reverse this trend through product iteration. DEXpert V2 is positioned as an all-in-one toolkit for decentralized exchanges on public blockchains. Its core component, BirdFly V1, is a dedicated launchpad focused on the creation and trading of meme tokens, offering token creation, liquidity migration tools, custom filters, and social media aggregation features. The logic behind this strategy is clear: meme token activity has been one of the most consistent drivers of trading volume in DeFi over the past two years. Additionally, DODO plans to add support for Solana and SVM blockchains, which will significantly expand its addressable market.

DODO’s challenges are equally apparent: meme token activity is highly cyclical and speculative, which can lead to significant volatility in platform utility. With PancakeSwap commanding 44% of DEX trading volume, DODO must strike a sustainable balance between its technical edge in capital efficiency and user acquisition.

The Inflection Point for Privacy Computing: FHE Moves On-Chain

On July 24, 2026, privacy computing project Zama announced the launch of its Confidential RFQ (Request for Quote) protocol, now live in private beta on Ethereum mainnet and scheduled for public release in September. The protocol is based on fully homomorphic encryption (FHE) technology, allowing on-chain concealment of trade direction, trade size, and slippage parameters to reduce MEV (miner extractable value) and frontrunning risks. The protocol enables market makers to compete via sealed-bid auctions and will soon support hidden asset types and multi-chain expansion.

This product launch comes on the heels of FHE’s leap from academic research to production-grade applications over the past 12 months. In June 2025, Zama completed a $57 million funding round at a $1 billion valuation, becoming the first FHE unicorn. In December 2025, its fhEVM went live on Ethereum mainnet. In January 2026, the ZAMA token was auctioned.

As of July 24, Gate market data shows the ZAMA token price at $0.05022, up 1.07% over 24 hours, 44.14% over seven days, and 62.16% over 30 days, with a market cap of about $122 million. Over the past week, ZAMA climbed from a low of $0.03407 to a high of $0.05970. Gate data shows ZAMA surged 28.59% on July 23, with a 24-hour trading volume of $321,000. Zama is committed to providing infrastructure for institutional-grade use cases such as confidential RWA tokenization, confidential payments, and confidential DeFi trading.

The value of FHE lies in its solution to the "shared state" privacy problem. Zero-knowledge proofs (ZKPs) excel at letting one party prove knowledge of a secret without revealing it, but cannot enable two users to interact with a shared encrypted state without disclosing information to each other. FHE, on the other hand, allows direct computation on encrypted data, making truly private smart contracts possible—the network never sees the underlying values. Zama’s architecture uses lightweight "handles" (ciphertext representations of encrypted values) to run smart contracts, while actual FHE computations are offloaded asynchronously to specialized coprocessors—so the on-chain layer never sees plaintext states.

Performance bottlenecks were once the biggest obstacle to FHE commercialization. Current benchmarks show Zama’s coprocessors can process over 20 transactions per second. GPU-accelerated FHE research indicates that using consumer-grade NVIDIA hardware, bootstrapping time is just 7.5 milliseconds. Zama’s technology is now post-quantum secure, with computation speeds over 100 times faster than five years ago.

Ethereum co-founder Vitalik Buterin wrote in January 2026 that this is the "year of reclaiming computational autonomy," recommending a fusion of ZKP, TEE, and FHE technologies to solve the "impossible triangle" of capability, cost, and privacy in local LLM deployment. This view reflects a key trend in privacy computing: the debate is no longer about which technology will win, but about how to combine the ZK, FHE, and TEE paradigms.

Intersection: The Fusion of Private Trading and Liquidity

Zama’s Confidential RFQ protocol sits squarely at the intersection of DEX liquidity and privacy computing. MEV and frontrunning have long plagued DEXs—traders who spot large unconfirmed orders in the mempool can profit by trading ahead, ultimately harming the average user’s experience. FHE technology addresses this at the cryptographic level by hiding trade direction, size, and slippage parameters on-chain.

The practical adoption of this technology could profoundly impact the DEX liquidity landscape. If Confidential RFQ effectively reduces MEV extraction, market makers will be more willing to provide deep on-chain liquidity without fear that their quoting strategies will be exploited. This could lower trading costs and tighten spreads at the DEX level, boosting overall trading efficiency—a technical parallel to Uniswap V4’s strategy of reducing LP incentives to cut trading costs.

Meanwhile, privacy computing is becoming essential infrastructure for institutional use cases. As RWA (real-world asset) tokenization advances, institutions need to protect business confidentiality while maintaining regulatory transparency. Zama’s FHE technology enables confidential assets to be natively issued, traded, and managed on public blockchains, with full verifiability and programmable compliance. This capability could become a key bridge between traditional finance and on-chain finance.

Conclusion

In July 2026, the Web3 infrastructure layer is evolving along two clear trajectories. In the DEX sector, Uniswap is leveraging its scale to pursue refined operations, aiming to drive growth by lowering trading costs. DODO, meanwhile, is capitalizing on the capital efficiency of its PMM algorithm, seeking new markets in meme tokens and the Solana ecosystem. In privacy computing, Zama’s Confidential RFQ protocol marks the entry of FHE technology into on-chain trading, offering cryptographic solutions to MEV and institutional privacy challenges.

Both evolutionary paths point to the same outcome: the Web3 application layer is gaining a more mature foundational base. DEX liquidity optimization addresses trading efficiency and capital utilization, while privacy computing tackles data sovereignty and compliance. As these two trends converge in on-chain trading scenarios, Web3 will deliver trading experiences closer to those of traditional finance—while offering privacy protections beyond the reach of legacy financial systems.

From a market size perspective, this remains an early-stage narrative. ZAMA’s market cap is $122 million, DODO’s is $19.61 million, and UNI’s is $2.358 billion—numbers that are negligible compared to the global financial system. But technological evolution is rarely linear. As FHE computation costs continue to fall and DEX capital efficiency keeps improving, programmable finance and privacy on-chain will unlock use cases previously unimaginable. For the infrastructure layer, 2026 could well be the year when incremental change reaches a tipping point.

FAQ

Q: What is the fundamental difference between FHE and ZKP in privacy protection?

ZKP (zero-knowledge proof) allows one party to prove a statement is true—such as "I have enough balance"—without revealing the actual value. FHE (fully homomorphic encryption) enables direct computation on encrypted data, so the network never sees the underlying data. In short, ZKP verifies truth, while FHE processes encrypted data.

Q: How does Uniswap V4’s reduction of LP incentives affect liquidity providers?

Uniswap proposes to cut V4 LP incentives by 33%, aiming to boost trading volume by lowering trading costs and tightening spreads. For LPs, short-term returns may drop, but if trading volume grows fast enough, total fee income could actually rise. The risk is that LPs may move their capital to competing protocols offering higher yields.

Q: How does DODO’s PMM algorithm differ from Uniswap’s AMM model?

Uniswap’s AMM uses a constant product formula (x × y = k) for automated pricing, while DODO’s PMM algorithm leverages price oracles to obtain market prices and concentrates liquidity near the market price. This allows DODO to offer deeper trading depth and lower slippage with the same amount of capital, resulting in higher capital efficiency.

Q: How does Zama’s Confidential RFQ protocol reduce MEV?

The Confidential RFQ protocol uses FHE technology to hide trade direction, size, and slippage parameters on-chain. Market makers compete via sealed bids, and trade details are invisible on-chain, preventing frontrunning bots from spotting and exploiting pending orders in the mempool.

Q: What are the core application scenarios for privacy computing in Web3 in 2026?

Key use cases include: confidential RWA tokenization (protecting business confidentiality for institutional assets on-chain), confidential DeFi trading (hiding positions and strategies), confidential payments and payroll (protecting both parties’ privacy), and confidential token distribution. As FHE computation costs decrease, these scenarios are moving from theory to production.