Crypto’s Infrastructure Unwind: L2 Shutdowns, Stablecoin Depegs, and Governance Alarms

The era of every project running its own chain is fading fast. In a single week, multiple protocols signaled that the cost of maintaining standalone infrastructure no longer adds up. A ZK-powered Layer 2 that raised $60 million is shutting down its blockchain, Berachain is rewriting its core incentive layer with a hard fork, and Kraken-incubated Ink is outsourcing chain operations to Optimism. Meanwhile, two overcollateralized stablecoins are trading well below their dollar pegs, a Tornado Cash DAO proposal raised alarms over a potential $23 million governance exploit, and Polymarket’s internal cash token revealed extreme wealth concentration. The picture that emerges from the weekly roundup from WuBlockchain is one of projects soberly cutting costs, retiring tokens, and refocusing on applications rather than infrastructure — a shift that could reshape how value flows across the ecosystem.

The End of Standalone Infrastructure?

Sophon’s announcement was the starkest. The ZK-powered Layer 2, which previously brought in $60 million in funding, told users it will shut down its native chain and migrate to Base under the new brand SOPH. Sophon stated plainly that the crypto infrastructure era is over, citing millions of dollars in annual maintenance costs that failed to generate enough value. It plans to launch Pyre, an “entertainment finance” payment app with gamified features, in early July. The decision reflects a broader calculation: launching a chain is cheap, but keeping it secure, liquid, and connected to users is an ongoing cost that few projects are covering.

Ink, the Ethereum L2 incubated by Kraken, chose a different path. It signed a multi-year agreement to move onto Optimism’s fully-managed OP Enterprise service. Optimism will operate Ink’s production infrastructure while the Ink Foundation focuses on ecosystem expansion and new financial products. The deal is one of the earliest instances of a major L2 outsourcing its stack to a managed provider. Ink’s roadmap still aims for aggressive technical targets — programmable block building, one-day withdrawal windows, 400 megagas per second of throughput, and 100-millisecond block times by end of 2026 — but the heavy lifting now sits with Optimism’s operations team.

Berachain’s July 8 hard fork marks a different kind of pullback: a token model reset. The Proof-of-Liquidity governance token BGT will be deprecated entirely, with BERA becoming the primary economic unit and sWBERA the value-accrual layer. Emissions will no longer flow through Boost voting. Instead, protocols must demonstrate real on-chain revenue and utility to qualify for emissions under the new routing mechanism ERAs. It’s a measured move away from the liquidity-mining era and toward a model that ties incentives to actual product usage, a theme rippling across the sector.

Stablecoins Under Pressure

Stablecoin markets rarely move quietly, and this week was no exception. Abracadabra Money’s MIM slumped to around $0.53, a 33% decline in 24 hours. That’s a searing depeg for an overcollateralized stablecoin. The team responded with emergency borrowing rate hikes across all Cauldron markets, betting that the discount would lure borrowers to buy MIM cheap and repay their debts, shrinking supply. Curve bribes and direct liquidity incentives were suspended until the peg recovers. The mechanism may work in theory, but with the token this far from its dollar target, the market is watching whether forced contraction can happen fast enough without triggering cascading liquidations.

Synthetix put forward SIP-423, a governance proposal to fully retire sUSD. The stablecoin trades around $0.25, a fraction of its intended $1 peg. Under the plan, sUSD holders would receive 4 SNX per 1 sUSD, but the compensation comes with a one-year lock-up and a second-year linear vesting period. The proposal, backed by founder Kain Warwick, amounts to an admission that the peg mechanism cannot be repaired. It also forces a choice: accept the swap and wait, or hold a deeply impaired asset. For a protocol that once anchored much of on-chain synthetic asset trading, the shutdown of its native stablecoin closes a significant chapter.

Stablecoin stress is not isolated. Across the broader tokenized asset market, capital continues to flow toward products with clearer institutional backing and regulatory pathing, leaving behind experimental designs that once flourished in zero-rate DeFi environments. What separates MIM and sUSD from safer dollar-pegged assets is the fragility of their collateral and redemption mechanisms under stressed conditions.

Governance and Concentration Risks

Lido’s governance vote to revoke official recognition for wstETH bridge endpoints across nine networks — including zkSync Era, Scroll, Mantle, and Polygon PoS — was a quiet but telling resource adjustment. The DAO will no longer actively monitor or support those deployments. The bridge contracts stay open and tokens remain valid, but the signal is clear: Lido is narrowing its attention to chains where liquid staking demand justifies the cost. That retreat may ripple into developer activity on those chains as staked ETH liquidity thins and composability with restaking and lending protocols weakens.

A more immediate threat surfaced at the Tornado Cash DAO. Researchers at L2BEAT flagged a malicious governance proposal containing an unverified target contract with complex decompiled logic. If passed, a delegatecall could compromise the DAO and its $23 million in TORN tokens. The proposer received funds through Railgun four days earlier, a privacy-preserving protocol that obscures the source. The core Tornado Cash protocol fund remains unaffected, but the incident is a reminder that DAO treasuries remain soft targets when voting participation is thin and proposal review is lax.

Polymarket’s on-chain cash balance token pUSD now holds a total supply above $500 million across over one million addresses, but the distribution is lopsided. Just 567 addresses — 0.06% of all holders — control 55.34% of the supply. Addresses with more than 1,000 pUSD make up 2.76% of holders and own 90.34%. Half of all addresses hold less than 10 pUSD. Some of that disparity reflects capital deployed in open prediction positions, but it still underscores how prediction market liquidity is concentrated among a handful of large traders. Thin participation at the retail level is a structural risk for platforms that rely on deep, diverse pools to produce accurate market signals.

BitGo’s announcement of a nearly 15% workforce reduction, refocusing on stablecoins, trading, security, settlement, and AI infrastructure, rounds out the week’s theme of consolidation. The custody firm is not exiting crypto; it is shedding non-core operations as the financial service landscape tightens. Whether this signals a broader cost-cutting wave across the sector or a one-off strategic pivot remains to be seen, but the direction is the same: attention is shifting from building infrastructure for its own sake toward products that generate sustainable revenue.