Cryptocurrency markets struggled broadly in the second quarter of 2026, with declines across stablecoins, spot trading and derivatives, but prediction markets reached record highs.

Spot trading volume across the top 10 centralized exchanges (CEXs) fell to $1.95 trillion in the second quarter of 2026, a 27.9% drop from $2.7 trillion in Q1, according to CoinGecko’s latest Crypto Industry Report published Thursday.

CEX perpetual futures volume also declined 10% to $12.7 trillion, while the stablecoin market slipped 1.6% to $305.1 billion. In contrast, prediction markets recorded their strongest quarter on record with $113.8 billion in notional volume.

The divergence highlights the growing role of prediction markets, with sports and politics emerging as the sector’s biggest drivers. Polymarket’s World Cup winner market alone has attracted more than $3.3 billion in trading volume, while contracts tied to the 2028 US presidential election rank among the platform’s largest markets, according to Polymarketscan data.

Source: Polymarketscan

Binance extends dominance despite bear market as DEX activity falls

Despite the bear market, Binance extended its dominance, with a 38.7% market share in Q2. In contrast, MEXC saw the biggest slump among spot CEXs, with trading volume more than halving from $275.2 billion in Q1 to $121.2 billion in Q2.

DEX activity also weakened during the quarter, with the top 10 spot DEXs processing $408.9 billion in volume, down from $556.4 billion in Q1. Uniswap strengthened its position as the leading DEX, holding a 41.2% market share despite a 21.4% drop in volume to $168.5 billion.

Source: CoinGecko

The declines came as the broader crypto market weakened, with total market capitalization falling 12.6% to $2.1 trillion during the quarter. April also marked a record month for hacks in decentralized finance (DeFi), highlighting ongoing security concerns across decentralized platforms.

Kalshi holds lead as prediction markets expand

Prediction market activity peaked in June, coinciding with the start of the FIFA World Cup, as monthly notional volume — the total value of all contracts traded — reached an all-time high of $50.7 billion, up 91.9% from the average of the previous five months.

Kalshi, the largest prediction market platform, maintained its lead over the quarter with a 58.9% market share, while Polymarket lost share from 35.8% to 30.2%. Robinhood-backed Rothera Markets climbed to fourth place.

Related: Czech Republic tells ISPs to block Polymarket after gambling blacklisting

The growth has drawn regulatory attention. In the US, regulators and states have clashed over whether prediction markets should be treated as financial markets or gambling platforms, with lawsuits involving platforms such as Kalshi escalating in 2026.

Authorities in other jurisdictions have also moved to restrict prediction markets, citing concerns including gambling rules, market integrity and potential insider trading risks.

Magazine: Has Bitcoin bottomed for this cycle? Analysts say ‘not yet’

Introduction

Credit is one of the most important pillars of the global economy. It enables individuals to buy homes, businesses to expand operations, and entrepreneurs to turn ideas into reality. Yet traditional credit markets remain slow, expensive, and often inaccessible to billions of people worldwide due to geographical limitations, rigid banking requirements, and outdated infrastructure.

Blockchain technology is transforming this landscape by introducing decentralized, transparent, and programmable credit markets. Instead of relying solely on banks and centralized financial institutions, blockchain enables borrowers and lenders to connect directly through smart contracts, creating a more efficient and inclusive financial ecosystem.

As decentralized finance (DeFi) continues to mature, blockchain-based credit markets are emerging as one of the most promising innovations capable of reshaping how capital flows across the globe.

What Are Blockchain-Based Credit Markets?

Blockchain-based credit markets are decentralized platforms where users can lend, borrow, or access credit using blockchain technology instead of traditional banking systems.

These platforms automate lending agreements through smart contracts, removing many intermediaries that typically increase costs and processing times.

Borrowers can access liquidity by providing collateral or, increasingly, through reputation-based lending models. Lenders earn yield by supplying digital assets into lending pools, where capital is allocated automatically based on transparent rules.

Everything—from interest calculations to loan repayments—is recorded on-chain, providing unprecedented transparency.

How They Work

A typical blockchain credit market follows a simple process:

1. Capital Providers Supply Liquidity

Investors deposit cryptocurrencies or stablecoins into lending pools.

2. Borrowers Request Loans

Users borrow assets by locking collateral or meeting protocol-specific credit requirements.

3. Smart Contracts Execute Loans

Instead of paperwork or manual approval, smart contracts automatically manage:

• Loan issuance
• Interest calculations
• Repayment schedules
• Liquidation rules
• Risk management

4. Lenders Earn Yield

Interest payments are distributed automatically to liquidity providers based on their contributions.

Why Blockchain Credit Markets Matter

1. Global Financial Inclusion

Traditional banks reject millions of borrowers due to geography, lack of documentation, or limited credit history.

Blockchain allows anyone with an internet connection and a crypto wallet to participate in global lending markets.

This opens financial opportunities for underserved populations worldwide.

2. Faster Loan Processing

Traditional loans can take days or even weeks.

Blockchain loans can be issued within minutes because smart contracts automate approvals and settlements.

3. Lower Costs

Removing intermediaries significantly reduces:

• Administrative fees
• Processing costs
• Cross-border transfer expenses
• Settlement delays

This benefits both borrowers and lenders.

4. Transparent Risk Management

Every transaction is recorded on-chain.

Investors can verify:

• Outstanding loans
• Pool liquidity
• Collateral levels
• Historical repayments

Transparency reduces information asymmetry that often exists in traditional finance.

5. Programmable Finance

Loans can include automated conditions such as:

• Dynamic interest rates
• Auto-repayment mechanisms
• Revenue-sharing agreements
• Milestone-based funding
• Tokenized collateral

This flexibility enables entirely new financial products.

The Rise of On-Chain Credit Scoring

One of the biggest challenges for decentralized lending has been undercollateralization.

Most DeFi loans today require borrowers to deposit more collateral than they borrow.

However, blockchain-native credit scoring is beginning to change this.

Protocols are developing reputation systems based on:

• Wallet history
• Repayment behavior
• On-chain transaction activity
• Governance participation
• Identity attestations
• Verifiable credentials

Instead of relying solely on traditional credit bureaus, future lending decisions may increasingly be based on verifiable blockchain activity.

Institutional Adoption Is Growing

Major financial institutions are beginning to recognize blockchain credit infrastructure.

Tokenized treasury products, real-world assets (RWAs), and on-chain lending protocols are creating bridges between traditional finance and decentralized finance.

Institutional lenders can now deploy capital more efficiently while benefiting from transparent risk monitoring and automated settlement.

As regulatory frameworks mature, institutional participation is expected to accelerate.

Challenges Still Ahead

Despite rapid innovation, blockchain credit markets face several obstacles:

Regulatory Uncertainty

Many jurisdictions are still developing rules for decentralized lending and digital assets.

Smart Contract Risks

Software vulnerabilities can expose lending protocols to exploits if not properly audited.

Identity and Reputation

Building secure, privacy-preserving decentralized identity systems remains an ongoing challenge.

Market Volatility

Crypto collateral can experience significant price fluctuations, increasing liquidation risk.

User Experience

Simplifying wallets, onboarding, and risk management remains essential for mainstream adoption.

The Future of Blockchain Credit

Several trends are likely to define the next generation of blockchain lending:

• AI-assisted credit risk assessment
• Zero-knowledge identity verification
• Cross-chain lending markets
• Tokenized real-world collateral
• Reputation-based unsecured lending
• Decentralized business financing
• On-chain corporate credit markets
• Embedded DeFi lending inside fintech applications

Over time, blockchain credit markets may evolve beyond crypto-native users and become foundational infrastructure for global digital finance.

Conclusion

Blockchain-based credit markets are redefining how capital is created, distributed, and managed. By replacing manual processes with transparent smart contracts, they make lending faster, more efficient, and more accessible to people around the world.

While challenges such as regulation, security, and identity remain, innovation is advancing quickly. As decentralized identity, tokenized real-world assets, and AI-powered risk assessment continue to mature, blockchain credit markets are poised to become a key component of the future financial system.

The evolution of credit is no longer limited to traditional banks. It is moving on-chain—where transparency, programmability, and global accessibility are creating a more open and inclusive financial ecosystem.

REQUEST AN ARTICLE

Ledger said is bringing its hardware security model to the fast-growing world of AI agents with the launch of Ledger Agent Stack, an open-source toolkit that allows autonomous software to interact with crypto wallets without ever controlling private keys.

The toolkit lets AI agents read wallet balances, analyze portfolios, prepare transactions and propose payments, but requires every sensitive action to be explicitly approved on a Ledger hardware device before it can be executed.

This is the first product release under Ledger’s 2026 AI roadmap, as the hardware wallet maker bets that human oversight will become a critical security layer as AI agents take on increasingly complex financial tasks.

“Agents propose. Humans approve,” the team wrote in their press release shared with CoinDesk. “Crypto wallets have protected billions on this standard for years,” said Ian Rogers, Ledger’s chief human agency officer, in the press release. “Ledger Agent Stack allows your agent to use these wallets just as easily as humans.”

President Donald Trump said in January he had no plans to pardon Bankman-Fried. He has cleared Binance founder Changpeng Zhao and Silk Road creator Ross Ulbricht, along with other white-collar offenders.

Bankman-Fried ran two companies at once. FTX was a crypto exchange, which holds customer money the way a broker does and is not supposed to touch it. Alameda Research was a trading firm he also owned. He moved billions of dollars in FTX customer deposits to Alameda, which spent the money on trades, venture investments, political donations, and Bahamian real estate, while FTX’s software exempted Alameda from the rules that would have forced it to cover its losses like any other trader.

The facade was blown open after CoinDesk obtained Alameda’s balance sheet in November 2022 and found that most of what the firm counted as assets was FTT – a token FTX had created itself and could issue at will.

The collateral propping up Alameda was, in effect, something its sister company had invented. Further cracks emerged after the prominent exchange Binance said, days later, it would sell its FTT holdings, leading to a rapid collapse in FTT prices.

Customers rushed to pull their deposits, and FTX could not return the money because it was no longer there. The exchange filed for bankruptcy on Nov. 11, 2022, just over a week after the story ran.

The global market capitalization of tokenized stocks rose to a record $2.3 billion on Wednesday, as more investors sought exposure to blockchain-based equity products.

The Ethereum network boasted the largest market share, at 34%, followed by BNB Chain with 30% and the Solana network with 23%, data aggregator Token Terminal shared in a Wednesday X post.

The largest increase came from Kraken exchange’s xStocks, which accounted for $507 million worth of tokenized stocks and Binance’s bStocks, with $334 million. Ondo Finance remained the largest tokenized stock issuer with $955 million in onchain equities, according to Token Terminal data.

More crypto platforms are bringing traditional investment products onto blockchain rails, in many cases opening up assets to investors outside of the United States. Tokenized stocks can also provide investors with fractional ownership and around-the-clock trading.

The world’s largest crypto exchange, Binance, opened zero-commission trading to over 7,000 US tokenized stocks for eligible users on June 1, as part of its strategy to become a multi-asset platform. Coinbase rolled out commission-free US stock and ETF trading with 24/5 availability in December 2025. 

Source: Token Terminal

In April, crypto exchange Bitget launched a proxy offering tied to the pre-initial public offering (IPO) phase of Elon Musk’s aerospace manufacturing and space transportation company, SpaceX.

In January, Vienna-based crypto exchange Bitpanda shared plans to expand its offering to include about 10,000 stocks and exchange-traded funds (ETFs). 

Kraken launched access to 11,000 US-listed stocks and ETFs through xStocks in April 2025 as one of the first major crypto-native platforms. The cumulative trading volume of xStocks exceeded $25 billion within about eight months of launch. 

Related: NYSE parent ICE pushes ‘level playing field’ for 24/7 onchain perps 

Tokenized RWAs surge despite crypto market pullback

The market for tokenized real-world assets (RWAs) surged 589% from early 2025 to June 2026, led by government bonds and money market funds, according to a June report from Binance Research.

Tokenized precious metals also attracted about $1.5 billion in value, rising 39% during the period.

The market for tokenized RWAs is becoming more diversified.
Source: Binance Research

However, stocks remain a small fraction of the broader tokenized RWA market, accounting for only about 5.5% of the $34 billion RWA market capitalization. 

Tokenized RWA market overview, all-time chart. Source: RWA.xyz 

About $15 billion in tokenized US Treasury debt represents the largest segment, or 44%, of the RWA market, followed by $4.5 billion in tokenized commodities, accounting for 13%, according to data provider RWA.xyz. 

Magazine: The 5 types of real world assets being tokenized fastest onchain

AI agents are starting to change how payments need to work, and Visa is warning that the world’s card rails were never built for the transaction volume, speed, and cost sensitivity required by machine-to-machine commerce. In a joint report released this week with investment thesis platform Artemis, Visa argues that AI-driven micropayments demand near-zero fees and faster settlement to become commercially viable.

The report links the timing to a capability shift: it says AI agents crossed a key threshold in mid-2025, enabling them to discover unfamiliar APIs, assess pricing, and make autonomous payment decisions. That capability, Visa and Artemis say, is pushing commerce toward an agentic model—while existing infrastructure gaps continue to limit mainstream adoption.

Key takeaways

• Visa and Artemis say traditional cards are optimized for low-frequency, human-paced payments, making them poorly matched to high-frequency AI micropayments.
• The report frames 2025 as a turning point for “agentic” payment automation, driven by AI agents’ ability to autonomously evaluate and execute transactions.
• Visa and Artemis propose convergence: cards remain useful for proxy purchases inside merchant networks, while stablecoins fit machine-native micropayments.
• Some machine-payment standards are already gaining momentum, including Coinbase’s x402 protocol, which the report cites as reaching sharply higher monthly transaction counts in late 2025.

Why AI agents stress existing payment rails

Visa and Artemis describe the underlying mismatch as structural. Cards were designed for human commerce—where transactions tend to be less frequent and costs can be amortized across larger purchase sizes. AI agents, by contrast, are expected to generate many more micro-transactions, often on short time horizons and with different cost constraints.

For agentic micropayments to scale, the infrastructure must support payments with extremely low per-transaction overhead and settlement that enables rapid decision-making loops. The report suggests that without these properties, the economics of autonomous “machine-to-machine” payments remain unfavorable—even if the AI itself can now carry out the payment decision.

The report also warns that infrastructure gaps aren’t just theoretical. It frames current limitations as a direct brake on adoption, arguing that the shift from experimenting with agent capabilities to using them in everyday transactions depends on payment networks that can keep up.

x402 shows demand for machine-native transaction patterns

While Visa and Artemis emphasize the broader need for new infrastructure, they also point to early signs that some agentic payment designs are gaining usage. The report cites the x402 payment protocol developed by Coinbase as an example of standards beginning to attract real transactional volume.

According to the report, x402 processed $15 million in adjusted volume across more than 109 million adjusted transactions since its May 2025 launch. It also highlights a significant acceleration in October 2025: the monthly transaction count reportedly increased from 40,000 to 3.8 million, resulting in 38 million transactions processed in October alone. The report attributes this momentum to the growing practicality of the protocol for machine-style payments.

Stablecoins as an on-ramp to agentic micropayments

A central claim in the Visa-Artemis report is that stablecoins could play a key role in enabling machine-native micropayments without forcing all commerce to migrate away from cards. Rather than positioning stablecoins and card networks as direct competitors, the report argues for a convergence model.

“The trajectory points toward convergence rather than competition: cards for proxy purchases inside existing merchant networks, stablecoins for machine-native micropayments, and hybrid flows where both are used within the same workflow.”

This framing matters for investors and builders because it implies that payment interoperability—not replacement—will likely define near-term strategy. In practice, an AI agent workflow may still need cards for certain merchant-bound transactions, but stablecoins (or other crypto rails) could be better suited for the agent’s “background” actions that require very low fees and fast settlement.

The report adds that a single machine-payments framework could support both stablecoin-based flows and traditional card transactions. Visa says this creates a path for card networks to plug into agentic payments over time, rather than treating agent commerce as a separate universe.

Protocols and tooling: connecting onchain and fiat through tokens

Visa and Artemis point to a broader effort to unify payment approaches using shared payment tokens and machine-payment protocols. The report says Tempo’s Machine Payment Protocol (MPP) now spans both onchain crypto payments and fiat payments through shared payment tokens.

Visa also states that its Card Specification SDK was built to extend the protocol into card-based agent commerce. It further notes that Tempo and Visa’s crypto division launched AI-related tools in March, with different focal points: Visa’s tooling is described as enabling same-day payments for AI agents, while Tempo’s Machine Payments Protocol is designed to make it easier for AI actors to send and receive money.

While the report doesn’t spell out new performance benchmarks for every component, it does make one clear point: adoption depends on interoperability across rails and token frameworks. If AI agents can consistently route payments through a machine-friendly layer, then the difference between stablecoin settlement and card settlement becomes a routing and workflow decision—not a structural barrier.

For readers watching the space, this suggests a practical trend: payment infrastructure will likely evolve through standards, SDKs, and shared protocol logic that allow agentic workflows to move between fiat and crypto settlement depending on cost, speed, and merchant coverage.

Going forward, the key question is whether agentic micropayment demand keeps accelerating fast enough to force mainstream payment networks to meet machine-level cost and settlement requirements. Buyers of payments infrastructure, developers integrating agent payments, and traders tracking adjacent stablecoin usage should watch how quickly token-based frameworks and machine payment standards—like x402 and MPP—translate early transaction growth into consistent, scaled deployments.

The implication does not stop there. According to Imran Lakha, founder of Options Insights, dealers hold a “net long gamma exposure” above $70,000. It means the dealers, who strive to maintain market-neutral exposure while making money from the bid-ask spread, would short or sell into strength above 70,000 to stay neutral or hedged.

“That hedging acts like a brake, capping how fast BTC can run once it gets up there,” Lakha said, adding that ether (ETH) isn’t as exposed to dealer gamma dynamics and can rip much faster.

Bitcoin was recently changung hands near $64,100, down nearly 1% since midnight UTC. Other major cryptocurrencies, including ether, XRP (XRP) and solana (SOL) nursed similar losses, while Nasdaq 100 index futures fell 0.5%.

“As always, there is a risk of a sudden sell-off amid financial market shocks, which could send BTC or global stock indices into a tailspin, but waiting for such moments is a thankless task,” said Alex Kuptsikevich, the chief market analyst at FxPro. “In such conditions, buying in a quiet market at less than half of peak levels looks like a perfectly reasonable tactic for the coming days or weeks.”

Stay alert!

Read more: For analysis of today’s activity in altcoins and derivatives, see Crypto Markets Today . For a comprehensive list of events this week, see CoinDesk’s “Crypto Week Ahead.”

Cointelegraph is committed to providing independent, high-quality journalism across the crypto, blockchain, AI, and fintech industries.

All news, reviews, and analyses are produced with full journalistic independence and integrity. For more details on our standards and processes, please read our Editorial Policy.