More than 100 crypto projects have shut down, filed for bankruptcy, or gone permanently dark in 2026, according to data compiled by RootData. The closures span every layer of the industry: centralized exchanges, Layer-1 and Layer-2 networks, DeFi protocols, NFT platforms, wallets, and infrastruct...
"Most of the projects now winding down were never generating revenue in the traditional sense. They paid engineers in tokens, subsidized liquidity in tokens and funded security audits in tokens. As long as those tokens held their dollar value, the system worked." — CoinDesk, August 2026
More than 100 crypto projects have shut down, filed for bankruptcy, or gone permanently dark in 2026, according to data compiled by RootData. The closures span every layer of the industry: centralized exchanges, Layer-1 and Layer-2 networks, DeFi protocols, NFT platforms, wallets, and infrastructure providers. Four major firms — BitMEX, BitMart, Movement Labs, and Storj Labs — announced closures or filings within a single week in late July.
The wave is structurally distinct from the 2022 collapse that brought down Terra, Celsius, and FTX through fraud and interconnected leverage. No single point of contagion exists this time. Instead, the industry faces a broad reckoning driven by three converging forces: altcoin prices down 70% to 90% from peaks draining token-denominated treasuries, $1.1 billion lost to on-chain exploits in H1 2026, and venture capital deal counts at five-year lows cutting off the rescue funding that kept marginal projects alive.
The survivors — protocols like Aave, Hyperliquid, and Uniswap that charge fees in stablecoins or cash — are consolidating market share. Six major protocols generated $726 million in revenue in H1 2026. The shakeout is shifting the market from speculative token distribution to proven business models, a pattern analysts have compared to the 2000-2001 dot-com bust.
RootData's 2026 Crypto Industry Dead Projects List documented 99 project closures through late July 2026. By early August, the count exceeded 100, according to CoinDesk and CryptoTimes reporting. The pace is accelerating: the week of July 21-27 alone saw BitMEX, BitMart, Movement Labs, and Storj Labs announce closures or bankruptcy filings.
The closures break down across sectors:
| Sector | Approximate Share of Closures | |--------|-------------------------------| | DeFi protocols | >50% | | Exchanges | ~10% | | Layer-1/Layer-2 networks | ~10% | | NFT platforms | ~8% | | Wallets and infrastructure | ~12% | | Other (gaming, analytics, etc.) | ~10% |
DeFi protocols account for more than half of all closures, according to CryptoTimes, which tracked 40+ DeFi-specific shutdowns through May 2026 alone. The remaining closures include centralized exchanges, blockchain networks, wallets (Leap Wallet among them), and infrastructure providers.
The 2026 shakeout differs fundamentally from the 2022 crypto winter. In 2022, fraud and interconnected leverage — Terra's algorithmic stablecoin, Celsius's rehypothecation, FTX's misuse of customer funds — created cascading failures. In 2026, the failures are almost entirely economic.
According to CoinDesk analysis published August 13, most shuttered projects "were never generating revenue in the traditional sense." The operating model was consistent across the casualties: pay engineers in project tokens, subsidize liquidity mining rewards in tokens, fund security audits in tokens. This model functioned only while token prices held their dollar value.
When altcoin prices declined 70% to 90% from cycle peaks, the mechanism collapsed. A project holding 10 million tokens valued at $1 each suddenly held a treasury worth $100,000 to $300,000 — insufficient to cover even a single quarter of engineering salaries.
Three primary failure modes emerged:
Treasury exhaustion. Projects that raised capital through token sales or venture rounds denominated in their native tokens saw purchasing power evaporate with price declines. Unlike cash-denominated raises, there was no floor.
Revenue absence. Most failed projects generated zero or near-zero fee revenue. They relied entirely on token appreciation and future fundraising rounds to sustain operations. When secondary market liquidity for mid-cap and small-cap tokens evaporated, the entire support structure disintegrated.
Security-driven insolvency. A subset of projects died not from business-model failure but from exploits that drained their treasuries or user deposits, making continued operation impossible. According to Blockaid, $1.1 billion was lost to 212 on-chain exploits in H1 2026 — more than all of 2025 combined.
Three centralized exchanges announced closures within July 2026, an unprecedented concentration.
BitMEX — the exchange that introduced 100x-leverage perpetual swaps to the crypto market — will cease operations on September 23, 2026. The Seychelles-based firm announced the closure on July 23 after what it described as a strategic review. HDR Global Trading Limited confirmed reserves exceed customer liabilities. The closure ends an 11-year run that defined an era of crypto derivatives trading. BitMEX halted new registrations immediately. From August 26, the exchange will block new positions and allow only position reductions before force-closing remaining trades.
BitMart announced its orderly wind-down on July 26 after nine years of operation. Trading services end August 26, with the platform fully closing January 31, 2027. The company cited operating conditions and market environment. Its BMX token fell 58% in 24 hours following the announcement.
AscendEX shut down on July 1, 2026, preceding the other two exchange closures.
None of the three exchanges described their closures as insolvency events. All three framed the decisions as strategic — a notable contrast to the 2022 exchange collapses, which were bankruptcy-driven.
The blockchain network layer has absorbed significant losses:
Movement Labs filed for Chapter 11 bankruptcy on July 21 in Delaware, listing under $500,000 in assets against up to $10 million in liabilities. The filing followed a scandal in which an intermediary connected to market maker Web3Port facilitated the sale of 66 million MOVE tokens — approximately 5% of supply — one day after the token's December 2024 launch. MOVE trades at roughly $0.011, down 99% from its $1.45 all-time high. Co-founder Rushi Manche, who separated from the company in May 2025, holds the largest creditor claim above $1.6 million and retains a 34.25% equity stake. Creditors have until September 14, 2026, to file claims.
BounceBit permanently shut down its Layer-1 blockchain on August 20 after an attacker exploited a protocol-level authorization vulnerability, extracting 286.5 million BB tokens (valued at approximately $3 million) from nine mainnet accounts. The team halted block production at height 20,702,857. BounceBit cited the discontinuation of the underlying Evmos project in May 2026 as making environment reconstruction unfeasible. The project is reissuing BB as a BEP-20 token on BNB Chain, effectively ending its existence as a standalone blockchain.
Loopring — Ethereum's first ZK-rollup DEX — shut down after years of declining adoption, citing weak user numbers, limited business development, and competition from newer zkEVM-based scaling networks. The team committed to distributing remaining user balances directly to Ethereum wallets, covering gas fees.
Swell sunset its Swellchain Layer-2 on June 15 to redirect resources toward a new product. Milkyway shut down its Celestia-based staking chain on March 26, citing low demand. Mint Blockchain, a real-world-asset Layer-1, ceased operations on April 17. ZKsync Lite halted block production on May 4 as part of a planned wind-down.
DeFi total value locked has fallen 39% in 2026, sliding from approximately $115 billion in January to roughly $70 billion, according to data aggregated by CryptoRank and DeFiLlama. The decline reflects both price depreciation and genuine capital withdrawal.
More than 40 DeFi protocols shut down through May 2026 alone, according to CryptoTimes. The failures cluster around specific protocol categories:
Lending protocols that relied on token incentives rather than organic borrowing demand saw utilization rates collapse. Without revenue, they could not sustain development or security audit costs.
Perpetual DEXs faced competition consolidation. Dango, a perpetual DEX and Layer-1, stopped trading July 29 and shut its chain by August 13 after concluding there was no path to profitability. Hyperliquid, by contrast, generated $871 million in revenue, effectively vacuuming volume from smaller competitors.
Yield aggregators and analytics platforms — including Goldfinch and Zapper — ceased operations as DeFi yields fell below traditional savings account rates. A CoinDesk report from April 2026 documented DeFi yields crashing to levels that could not compete with a simple bank savings account, undermining the sector's fundamental value proposition for retail users.
Ethereum, which leads DeFi market share, saw its DeFi TVL drop 43% to $38.91 billion, though it maintained a 53.1% share of total DeFi TVL, according to Coinlaw.
The Blockaid H1 2026 Security Report documented 212 verified on-chain exploits totaling $1.1 billion in losses — a record for any six-month period and exceeding all of 2025 combined. April 2026 was the most-hacked month in crypto history by number of incidents.
Four incidents accounted for approximately $707 million, or 64% of total H1 losses:
| Exploit | Amount | Date | |---------|--------|------| | KelpDAO | $292 million | April 18 | | Drift Protocol | $285 million | April 1 | | Resolv | ~$65 million | H1 2026 | | CowSwap | ~$65 million | H1 2026 |
Blockaid attributed the KelpDAO, Drift Protocol, and Humanity Protocol exploits to TraderTraitor, a North Korean state-sponsored subset of the Lazarus Group. These three attacks totaled roughly $609 million — 55% of all H1 losses.
Privileged key misuse was the costliest attack vector, accounting for approximately $790 million, or three-quarters of all stolen funds. Cross-chain bridges and EVM Layer-2 exploits remained persistent vulnerabilities.
The security losses have a direct connection to project closures. When a protocol loses its treasury or user deposits to an exploit, continued operation becomes impossible. Unlike well-capitalized firms that can absorb losses, most mid-tier DeFi protocols operate with minimal reserves.
Crypto venture funding has contracted sharply, removing the rescue financing that historically kept marginal projects alive.
Monthly VC deal count hit approximately 50 in May 2026, a five-year low, according to Yellow Research. Total capital deployed remained elevated but compressed into fewer, larger bets — a structural shift favoring established players over early-stage experiments.
Key metrics illustrate the severity:
The capital that remains is flowing to three categories: infrastructure projects, AI-adjacent protocols, and stablecoin-adjacent fintech. According to CoinDesk's August 15 analysis, $11.2 billion in 2026 funding was concentrated in these sectors, effectively starving generalist crypto projects.
The pattern mirrors what happened in the 2000-2001 dot-com bust: overall sector funding declined, but the money that remained concentrated in the eventual winners — Amazon, Google, eBay — while hundreds of funded competitors died.
The surviving protocols share a common characteristic: they generate fee revenue in stablecoins or fiat-denominated assets, not in their own tokens.
Six major protocols — Aave, Aerodrome, Hyperliquid, Pump, Sky, and Uniswap — generated $726 million in combined revenue in H1 2026. These protocols function as financial infrastructure with measurable unit economics.
Hyperliquid leads with $871 million in annualized revenue, driven by perpetual contract trading fees, spot market fees, builder code auctions, and HyperEVM gas fees. Its revenue model does not depend on token price appreciation.
Aave is projected to generate approximately $60 million in protocol profit in 2026, according to Grayscale estimates, with an operating margin of approximately 50%. Revenue derives from loan interest, flash loan fees, liquidation penalties, and GHO stablecoin interest.
Uniswap continues to generate revenue from swap fees across its deployments, maintaining its position as the dominant decentralized exchange.
Grayscale identified Aave, Uniswap, and Hyperliquid among the most undervalued crypto assets in 2026, based on the disconnect between their revenue generation and token prices. The implication: markets are pricing all crypto projects — revenue-generating and revenue-absent — through the same bearish lens, creating a potential mispricing.
The distinction between survivors and casualties aligns with a framework that values economic utility over speculative narratives. Projects that facilitate real transactions, charge fees for measurable services, and maintain sustainable cost structures are weathering the downturn. Projects that relied on token appreciation as their business model are not.
The 2026 crypto shakeout represents the industry's first purely economic reckoning. There is no villain — no Do Kwon, no Sam Bankman-Fried — to blame. The projects dying were, for the most part, legitimately attempting to build products. They failed because their business models depended on perpetually rising token prices, and those prices fell.
The comparison to the dot-com bust is structurally appropriate. Between 2000 and 2002, 48% of dot-com companies ceased operations, according to research published by the University of Maryland. The companies that survived — Amazon, Google, eBay — were distinguished by actual revenue and defensible market positions, not by the novelty of their technology. Crypto appears to be following the same script.
What remains after the shakeout is a smaller, more concentrated industry. Fewer exchanges, fewer Layer-1 and Layer-2 networks, fewer DeFi protocols — but the survivors generate measurable revenue. Whether this consolidation ultimately strengthens the industry's long-term economic foundation or simply reduces competition depends on whether the survivors use their market position to build sustainable value or merely extract rents from a smaller user base.
The data is clear on one point: the era of surviving on token appreciation alone is over.