Introduction
I have been in or adjacent to the Web3 space for years, and for the past roughly 18 months I have done almost no development work in it because the demand is not there. That is a statement of fact about my own pipeline. I am not predicting anything or lodging a complaint, and I hear the same thing from enough peers to believe my experience is not unique. When I sit down to write about trends for 2026, the honest question I have to address before anything else is whether Web3 as a concept is actually still a thing. By Web3 I mean the user-owned, decentralized, token-incentivized successor to Web2 that we were all promised around 2021. My view, and I will defend it below with data, is that the infrastructure layer is more alive than ever while the consumer Web3 narrative has largely collapsed. Some projects with real fundamentals are still going strong. Most of what got called Web3 was hype wrapped around token launches, and it did not survive contact with real users.
What Actually Happened to Web3 in 2025
The most direct measure of an open-source software movement is the number of developers actually working on it. That number collapsed in 2025. According to Artemis Terminal data, weekly active Web3 developers tagged in open-source repositories dropped from 12,380 in March 2024 to around 7,600 in March 2025, a 38.6% decline in a single year. [1][2] Total weekly commits across public crypto repositories have fallen around 75% since early 2025, from roughly 850,000 to about 210,000. [28] Ethereum remained the single largest share of that activity even as the absolute numbers shrank. [25][28]
A 38.6% drop in active developers in one year is what an exodus looks like. It is also happening while AI absorbs a significant share of technical talent. [3] Optimism contributor Binji Pande was direct about it, describing the drop as one of the clearest signals of long-term health and pointing out that attention has shifted, incentives have dried up and speculation has moved faster than utility. [2] He added that crypto needs to feel futuristic again if it wants to win builders back. [2]
By early 2026 the contraction had deepened rather than reversed. Artemis data cited by CoinDesk put weekly active crypto developers at roughly 4,600, down about 56%, with Ethereum's own weekly active developer count down about 34% over three months to around 2,800, even as GitHub added some 36 million developers in 2025, most of them building AI. [28]
Web3 gaming is the most brutal case study. In Q2 2025 alone, daily active wallets in Web3 gaming dropped 17% quarter-over-quarter, and funding collapsed 93% year-over-year to $73 million, the lowest in two years. [4] Over 300 gaming dApps went inactive. [4][5] High-profile shutdowns during 2025 included Deadrop (Midnight Society), Ember Sword (Bright Star Studios), Nyan Heroes, ChronoForge and Aether Games, all citing some combination of unsustainable tokenomics, inability to retain players after reward incentives stopped and funding shortages. [5][6] Aether Games raised $12.62 million and still failed. [7] The pattern was consistent. The games were not good enough to keep players once the token rewards ended.
This is the heart of what failed. The play-to-earn model treated games as yield extraction vehicles rather than games. When token prices fell, players left, which meant fewer players buying tokens, which meant more token price decline, which meant more players leaving. It was a predictable economic failure mode, and it happened at scale. By 2025, venture funding for Web3 gaming had fallen to roughly $360 million, down from about $4 billion in 2022, and gaming's share of crypto venture capital had dropped from 62.5% to single digits as money rotated into AI, real-world assets and infrastructure. [24] By April 2026, research firm Caladan estimated that more than 90% of Web3 games had failed after a boom that pulled in somewhere between $12 billion and $15 billion, because the gamers never showed up. [24]
What Did Not Fail
The honest caveat is that a lot did survive, and some of it is growing. DeFi total value locked sits around $160 billion as of mid-2026, recovering from the 2022 trough of roughly $52 billion though still below 2021 peaks. [8] That recovery was not a straight line. When a broader crypto selloff hit in February 2026, with Bitcoin sliding from around $96,000 in late January to briefly below $61,000, DeFi TVL fell only about 12%, from $120 billion to $105 billion, far less than the rest of the market, which suggests the remaining DeFi users are stickier and more yield-driven than speculative. [9] TVL bottomed near $90 billion in early February before climbing back to roughly $160 billion by June. [8][9] Institutional capital has moved toward regulated, on-chain finance, though the verifiable figures are more modest than some industry trackers suggest. DefiLlama data compiled by DL News put total crypto fundraising in 2025 above $25 billion, with the top ten raises alone bringing in more than $10 billion. [10]
Layer 2 infrastructure delivered. Layer 2 TVL grew from under $4 billion in 2023 to roughly $48 billion across 73 active rollups by April 2026, and daily L2 transactions exceeded 1.9 million through 2025. [11][29] Ethereum mainnet gas fell from 7.141 gwei in January 2025 to roughly 0.50 gwei in January 2026, a 93% decrease, and L2 fees routinely settle below one cent. [12] Ethereum's Merge reduced the network's energy consumption by approximately 99.95% on the Ethereum Foundation's pre-Merge estimate [13], with a later measurement by the Crypto Carbon Ratings Institute putting the reduction closer to 99.99%. [14] Either way, it is one of the cleanest wins in the history of blockchain technology and one that is no longer seriously disputed.
Stablecoins crossed from niche crypto primitive to genuine financial instrument. Combined stablecoin supply grew from $138 billion in February 2024 to $225 billion in February 2025, a 63% increase in a year, and has since climbed to a record of roughly $320 billion to $322 billion by May 2026, its fourth consecutive month of expansion even as broader digital asset prices trended down. [15][8] At that level the stablecoin market is now worth more than the foreign exchange reserves of 95 countries, including the United Kingdom and Canada. [8] The GENIUS Act, signed into US federal law on July 18, 2025, gave payment stablecoins their first comprehensive federal regulatory framework, though its core prohibitions do not take full effect until early 2027. [16] In the EU, MiCA has been in full application since December 30, 2024. More than 40 crypto-asset service provider licenses had been issued across member states by late 2025, the transitional grandfathering period runs out on July 1, 2026, and breaches can draw administrative fines of up to €5 million or 5% of annual turnover. [17][18]
Real-world asset (RWA) tokenization grew roughly 266% across 2025 and has since passed $32 billion in tokenized value by May 2026. [19][26] BlackRock's BUIDL fund alone accounts for about $2.3 billion to $2.5 billion. [26] Major banks are piloting tokenized bonds, with UBS and the European Investment Bank among the issuers. [31] JP Morgan's Kinexys has processed over $3 trillion in transactions since inception and now averages more than $5 billion a day. [20] Coinbase announced stock and ETF trading at its December 2025 product event and opened it to all eligible US users in February 2026, but its in-house tokenized equities, built on infrastructure it calls Tokenize, were still not live in the US as of mid-2026. [21][30] As of June 2026 Coinbase was launching those tokenized stocks in eligible non-US jurisdictions first, with no US date, after the SEC pulled its planned May 2026 innovation exemption for tokenized stocks when the major exchanges objected. [30][32]
So What Is Web3, Exactly, in 2026?
This is where the honest answer gets uncomfortable for the industry. What is growing is not really what most people meant by Web3 when they used the term in 2021 or 2022.
The original Web3 promise was a user-owned, decentralized alternative to the Web2 platform stack. That meant decentralized social media, user-controlled identity, open gaming economies where players owned their assets, peer-to-peer finance without intermediaries and user-owned data. That vision, as a mass-market proposition, has not materialized. Decentralized social media has not replaced Twitter/X or Instagram. User-owned identity is being rolled out, though through state-issued wallets like the EU Digital Identity Wallet rather than blockchain-native systems. [22] Web3 gaming failed to retain players. DeFi works, though it serves yield farmers and institutions rather than a new class of empowered retail users.
What did grow is the infrastructure layer. That includes regulated stablecoins, tokenized Treasuries, Layer 2 scaling, institutional RWA platforms and compliance-integrated crypto exchanges. All of this is better described as crypto rails absorbed into traditional finance than as Web3 in the 2021 sense. JP Morgan using blockchain for settlement does not qualify as Web3. BlackRock running a tokenized money market fund on Ethereum does not qualify as Web3. Coinbase preparing to sell you tokenized Apple stock sits closer to brokerage with a blockchain backend than to any decentralized future.
This is why I say, and my colleague is more or less right, that if you strictly define Web3 as the consumer-facing, decentralized, user-owned internet we were promised, most of that was hype and it did not deliver. The developer exodus, the gaming collapse, the failure of decentralized social platforms to break through and the absorption of self-sovereign identity by state-issued identity wallets together tell a consistent story. What was sold as a revolution became a piece of financial plumbing. Useful plumbing. Still plumbing.
The Stuff That Does Still Work
To be fair to the technology, the parts that kept their core principles are still going. Ethereum works, is energy-efficient and settles meaningful value. [13] In my view Bitcoin is still the most secure permissionless settlement system we have, and its core proposition of censorship-resistant, non-governmental money has not changed. Permissionless DeFi protocols like Aave, Uniswap and Maker (now Sky) continue to operate with meaningful TVL, even as they now compete with compliant institutional products. [23]
Self-custody still works. The cryptographic primitives, including signatures, zero-knowledge proofs, verifiable credentials and smart contract composability, are real technology with real applications. Privacy-preserving computation has advanced. Cross-chain interoperability has improved. The infrastructure built during the hype cycle is actually usable, and in many cases the people who stayed are doing the good work precisely because they care about the technology rather than the token price.
When someone asks me whether Web3 is still a thing, I separate that question into two parts.
As a consumer-facing, decentralized, user-owned alternative internet, mostly no. Most of that was marketing. The developer exodus, the gaming collapse and the failure of decentralized social alternatives are the data.
As technical infrastructure that has been absorbed into finance, identity and payments, yes. It is bigger and more regulated than ever. Most of it no longer calls itself Web3, though. It calls itself tokenization, stablecoin payments or on-chain settlement, and it is backed by institutions.
Why I Have Not Been Building
For my own work, the implication is straightforward. For about 18 months, there has been no real demand for classic Web3 development such as consumer dApps, NFT marketplaces, play-to-earn games or decentralized social features. The clients asking for those projects in 2022 and early 2023 are no longer asking in 2025 or 2026. What is being asked for is tokenization of specific financial instruments, stablecoin payment integrations, compliance tooling and increasingly agent-to-agent payment rails. Those projects exist, they pay and they are much closer to fintech development than to the Web3 culture of five years ago.
I care about the core principles of open networks, censorship-resistant settlement, user-controlled keys and programmable money. I do not think those principles are dead. I think they have been quietly absorbed into regulated infrastructure, which is simultaneously a win for the technology and a loss for the decentralized ideology that made it interesting in the first place. Both things can be true.
Conclusion
So, is Web3 still a thing? A narrow, honest answer: the infrastructure is thriving, the consumer narrative has collapsed and most of what survived no longer uses the word. The developer exodus is real. [1][2][3][28] The gaming failure is documented. [4][5][24] DeFi works, though mostly for institutions and yield seekers. [10] Digital identity is being deployed by governments rather than by Web3 platforms. [22] Stablecoins and RWA tokenization are growing fast, though they represent traditional finance using blockchain rails rather than blockchain replacing traditional finance. [19]
For anyone still working in or around the space, I think the honest path forward is to separate what you actually believe in from what the market rewards. The technology is good. The principles are defensible. The marketing was overblown. If you build with that clarity, you will find real work to do. If you keep waiting for the 2021 Web3 consumer wave to come back, you will be waiting for a while.
Sources
[1] Fully Crypto, Number of Active Web3 Developers Shrinks by 40% in a Year, April 2025, https://fullycrypto.com/number-of-active-web3-developers-shrinks-by-40-in-a-year.
[2] Cointelegraph, Web3 active developers drop nearly 40% in one year, April 2025, https://cointelegraph.com/news/web3-developer-activity-drop-nearly-40-percent-one-year.
[3] Traders Union, Blockchain loses engineers: Why developers exit Web3, May 2025, https://tradersunion.com/news/editors-picks/show/281656-blockchain-loses-engineers/.
[4] CoinDesk, Market Data Shows a Brutal Shakeout is Underway in Web3 Gaming, July 2025, https://www.coindesk.com/web3/2025/07/11/web3-gaming-faces-ongoing-turmoil-market-metrics-reveal-persistent-decline.
[5] GAM3S.GG, Is Blockchain Gaming Dead?, November 2025, https://gam3s.gg/news/is-blockchain-gaming-dead/.
[6] CCN, Did Web3 Gaming Die in 2025? The Five Reasons Behind GameFi's Decline, December 2025, https://www.ccn.com/news/crypto/web3-gaming-die-in-2025-the-five-reasons-behind-gamefi-decline/.
[7] PlayToEarn, Aether Games Shuts Down After Failing to Scale in Web3, December 2025, https://playtoearn.com/news/aether-games-shuts-down-after-failing-to-scale-in-web3.
[8] DefiLlama, DeFi TVL and stablecoin supply dashboards, accessed June 2026, https://defillama.com/.
[9] CoinDesk, DeFi's value holds up despite crypto sell-off as yield seekers stay put, February 2026, https://www.coindesk.com/business/2026/02/03/defi-s-quiet-strength-tvl-holds-as-market-selloff-tests-traders.
[10] DL News (DefiLlama data), Top 10 crypto fundraising rounds in 2025, 2026, https://www.dlnews.com/articles/markets/top-10-crypto-fundraising-rounds-in-2025/.
[11] Cryptopolitan, Layer 2 Adoption 2026 Predictions, December 2025, https://www.cryptopolitan.com/layer-2-adoption-2026-predictions/.
[12] BlockEden, Ethereum Layer 2 Solutions in 2026: Arbitrum, Optimism, and zkSync Head-to-Head, February 2026, https://blockeden.xyz/blog/2026/02/18/ethereum-layer-2-solutions-comparison-arbitrum-optimism-zksync/.
[13] Ethereum Foundation, The Merge, https://ethereum.org/roadmap/merge/.
[14] Consensys, Ethereum Blockchain Eliminates 99.99% of its Carbon Footprint Overnight After a Successful Merge According to New Report, September 2022, https://consensys.io/blog/ethereum-blockchain-eliminates-99-99-of-its-carbon-footprint-overnight-after-a-successful-merge-according-to-new-report.
[15] Cointelegraph, Stablecoin users grew 53% in one year, March 2025, https://cointelegraph.com/news/stablecoin-users-53-percent-growth-2025.
[16] The White House, Fact Sheet: President Donald J. Trump Signs GENIUS Act into Law, 18 July 2025, https://www.whitehouse.gov/fact-sheets/2025/07/fact-sheet-president-donald-j-trump-signs-genius-act-into-law/.
[17] Skadden, MiCA Update, Six Months in Application, July 2025, https://www.skadden.com/insights/publications/2025/07/mica-update-six-months-in-application.
[18] Cyfrin, MiCA Regulation Explained: A Guide To EU Crypto Compliance, November 2025, https://www.cyfrin.io/blog/mica-regulation-explained-a-guide-to-eu-crypto-compliance.
[19] Investax, Real World Asset Tokenization: Trends and Outlook for 2026, 2026, https://investax.io/blog/real-world-asset-tokenization-trends-and-outlook-for-2026.
[20] J.P. Morgan, Latest milestones at Kinexys by J.P. Morgan, April 2026, https://www.jpmorgan.com/payments/newsroom/kinexys-milestones-2026.
[21] CNBC, Coinbase adds prediction markets and stock trading in push to be one-stop trading app, December 2025, https://www.cnbc.com/2025/12/17/coinbase-prediction-markets-stock-trading-stablecoins.html.
[22] European Commission, European Digital Identity, March 2026, https://commission.europa.eu/topics/digital-economy-and-society/european-digital-identity_en.
[23] CoinLaw, Decentralized Finance (DeFi) Market Statistics 2026, February 2026, https://coinlaw.io/decentralized-finance-market-statistics/.
[24] CoinDesk, More than 90% of Web3 games failed after $15 billion boom as gamers never showed up: Caladan, April 2026, https://www.coindesk.com/markets/2026/04/23/more-than-90-of-web3-games-failed-after-usd15-billion-boom-as-gamers-never-showed-up-caladan.
[25] Cryptopolitan, Crypto developer activity returns to 2018 levels, Ethereum retains a significant share of innovation, May 2025, https://www.cryptopolitan.com/crypto-developer-activity-2018-ethereum/.
[26] MEXC Crypto Pulse (RWA.xyz data), RWA Tokenization in 2026: How to Invest in Real-World Assets on the Blockchain, May 2026, https://www.mexc.com/crypto-pulse/article/rwa-tokenization-in-2026-113521.
[27] AInvest, Web3 Developers Decline 38.6% in a Year, April 2025, https://www.ainvest.com/news/web3-developers-decline-38-6-year-2504/.
[28] CoinDesk, Crypto developer activity sinks to multi-year low as AI absorbs GitHub's talent boom, March 2026, https://www.coindesk.com/tech/2026/03/12/crypto-developer-activity-sinks-to-multi-year-low-as-ai-absorbs-github-s-talent-boom.
[29] L2BEAT (via Eco), Best Ethereum L2s in 2026: Fees, TVL, TPS Compared, April 2026, https://eco.com/support/en/articles/14798699-best-ethereum-l2s-in-2026-fees-tvl-tps-compared.
[30] Coinbase, System Update: The future of finance is on Coinbase, December 2025, https://www.coinbase.com/blog/system-update-the-future-of-finance-is-on-coinbase.
[31] MetaMask, RWA categories in 2026: tokenized Treasuries, equities, credit, and more, 2026, https://metamask.io/news/types-of-tokenized-real-world-assets-rwa-categories.
[32] CoinDesk, Coinbase joins tokenized stock race with onchain shares and dividend payments, June 2026, https://www.coindesk.com/business/2026/06/16/coinbase-to-join-tokenized-stock-race-with-onchain-shares-dividend-payments.
