Crypto is experiencing an unprecedented surge in attacks, not because the technology is getting weaker, but because artificial intelligence has made hacking economically viable at a scale never seen before. Through the first half of 2026, the industry recorded between 164 and 212 separate exploits, more than double the prior year's pace. Yet the total dollars stolen actually declined compared to 2025, revealing a shift in the nature of the threat itself.

What's Driving the Explosion in Crypto Hacks?

The numbers tell a striking story. By CoinGecko's count, 2026 logged 164 separate incidents through early August, already exceeding any full prior year, with the previous record being 2025's 97 incidents. TRM Labs, a blockchain security firm, recorded 207 hacks in just the first half of 2026, more than double the 83 from the same period a year earlier. Blockaid independently verified 212 exploits during the same window.

The dollar figures, however, paint a different picture. First-half 2026 losses came in around $972 million to $1.1 billion, below the $1.5 billion stolen in the same period of 2025. But that comparison is misleading. As TRM Labs analyst Ari Redbord noted, the year-over-year decline happened almost entirely because North Korea did not repeat an operation on the scale of the $1.5 billion Bybit hack from 2025. When that single outlier is removed, the trend becomes clear: more attacks, spread across more protocols, each taking less money.

This specific pattern points to a single cause. If attacks suddenly became cheaper to run, you would expect many more of them, reaching down to targets that were previously too small to bother with. That is exactly what the data shows. TRM reported in August 2026 that artificial intelligence adoption across crypto crime rose 40 percent year over year. The mechanism is straightforward: AI did not invent new crimes, it removed the constraints on old ones. The skill floor dropped, the scale ceiling lifted, and fake identity went industrial.

"It used to be too costly to hack someone worth $20,000, because the time was not worth it, and an AI agent can now go after everyone at once," explained Ryan Kirkley, Global Settlement Network, speaking at the Wyoming Blockchain Symposium.

Ryan Kirkley, Global Settlement Network

How Are Small Crypto Teams Being Overwhelmed?

The real-world impact of AI-assisted attacks is already visible across the crypto ecosystem. In August 2026, two Bitcoin swap services shut down within weeks of each other, both citing the same cause: relentless, automated probing they could not defend against.

Boltz, a Bitcoin swap service, suspended operations after describing months of steadily rising automated, AI-assisted probing that its small team could not patch fast enough. Atomiq followed shortly after, taking its swap routes offline because, as a small team, it could not fight the numerous sophisticated AI-assisted attacks on its infrastructure. These were not targeted attacks on specific vulnerabilities; they were waves of automated reconnaissance and exploitation that overwhelmed the teams' ability to respond.

The attacks are taking multiple forms across the industry. On August 23, Term Labs lost about $8.5 million through a governance exploit where no code was broken at all. The attacker simply acquired governance tokens, which cost a few dollars in vault shares, then held 100 percent of the vote on five of the drained vaults. He opened a proposal styled to look like a routine parameter update, waited out the six-day minimum voting period, and executed a bundle of 17 actions that recalled every asset into a strategy contract he controlled.

On August 22, Blockaid detected an ongoing exploit of The Sandbox's SAND token on Base, where attackers hijacked LayerZero delegate permissions and minted unbacked SAND across hundreds of transactions. The mechanism was a permissions oversight, but the automation is what made it relentless and profitable at scale.

Steps to Understanding Crypto's New Attack Landscape

• Volume Over Value: The 2026 trend shows attackers targeting many smaller protocols rather than attempting one massive heist, making the ecosystem feel less stable even when total losses are lower.
• Human Layer Vulnerability: The two largest first-half losses, Drift at roughly $285 million and KelpDAO at roughly $292 million, both traced to LinkedIn social engineering leading to a compromised multisig signer, the same human-layer attack that hits banks and enterprises.
• Operational Risk Rising: Small teams managing open-source infrastructure with liquid value are now the primary targets, not because their code is weaker, but because they lack the resources to defend against AI-assisted waves of attacks.

Is Crypto's Core Technology Actually Getting Weaker?

The encouraging news is that crypto's core smart-contract security has actually been improving, not decaying. Immunefi's six-year data shows DeFi protocol losses fell about 80 percent from the 2022 peak of $2.62 billion to $534 million in 2024, with the median loss per incident dropping from $6 million to $1.5 million even as total value locked grew substantially. The old ecosystem-class attacks, flash-loan oracle manipulations and reentrancy, collapsed from nearly 19 percent of losses in 2022 to under 1 percent in 2025.

What changed in 2026 is not that the code got worse. It is that AI made probing every layer, especially the human and operational layers, cheap enough to do at scale. That pressure is arriving everywhere software runs. Crypto simply feels it first because its infrastructure is open-source, its value is liquid, and its teams are often small.

The Bybit hack that defined 2025 was not a flaw in Ethereum or any blockchain protocol. It was a compromised interface at a wallet infrastructure provider. Similarly, the largest losses in 2026 trace to social engineering and operational failures, not cryptographic weaknesses. This distinction matters because it means the underlying technology is sound; the vulnerability is in how humans and organizations manage it.

What Defense Technology Is Actually Working?

The encouraging half of the story is that the capability driving the attacks is also the strongest available defense. The important qualifier is that this defense is not something a protocol can simply switch on, and that is by design.

Anthropic launched Project Glasswing on April 7, 2026, on a deliberate premise. The company had built a frontier artificial intelligence model, Claude Mythos, that it assessed could surpass all but the most skilled humans at finding and exploiting software vulnerabilities. Releasing that openly would hand the same capability to attackers, so Anthropic did the opposite and distributed it narrowly, to defenders of software whose compromise would be catastrophic.

The launch cohort was around 50 organizations and reads like a list of the world's most critical infrastructure: Apple, Microsoft, Amazon, Google, NVIDIA, JPMorgan Chase, Cisco, CrowdStrike, and the Linux Foundation among them. Access is invitation-only with no self-serve signup, and every organization has to meet Anthropic's security requirements before it is granted the model.

The early results were substantial. In roughly a month, Glasswing partners used the model to find more than 10,000 high- or critical-severity vulnerabilities across systemically important software, including a critical flaw in a cryptographic library used by billions of devices, since patched. One partner bank used it to detect and stop a fraudulent $1.5 million wire transfer. In May the program expanded to reach additional organizations.

The frontier of AI-assisted attacks is already visible in emerging threats. Researchers disclosed JadePuffer, described as the first fully agentic ransomware, where an AI agent ran reconnaissance, credential theft, lateral movement, and encryption end to end. Blockaid flagged a $216,000 exploit of an AI trading agent as the first of its kind, with researchers expecting prompt-injection attacks on agents to grow through the year.

The 2026 crypto hack surge is not a sign that blockchain technology is failing. It is a sign that artificial intelligence has fundamentally changed the economics of cybercrime, making small-scale attacks profitable for the first time. The same technology is now the strongest defense available, but access to it is carefully controlled to prevent it from becoming a tool for attackers. For crypto teams and protocols, the lesson is clear: the vulnerability is no longer primarily in the code, but in the human and operational layers that manage it.