TL;DR: April 2026 became the worst month on record for crypto security, with roughly 30 incidents and more than $625 million stolen across DeFi protocols, exchanges, bridges, and wallets, according to FYEO and DefiLlama. The pattern points to operator compromise, not contract failure, as the dominant loss driver, making key governance and signer trust the primary control problem.
At a glance
What this is: NHI Mgmt Group's analysis, drawing on FYEO's blog post: April 2026 set a new record for crypto losses, with attacks driven mainly by compromised keys, phishing, and weak signing processes rather than contract exploits.
Why it matters: For IAM, PAM, and NHI practitioners, the lesson is that signing authority, key custody, and human approval paths now function as the real control plane in crypto operations.
By the numbers:
- Roughly 30 separate incidents drained more than $625 million in April 2026, eclipsing every prior monthly total tracked on DefiLlama's Hacks dashboard.
- Private key compromises accounted for 22.3% of losses, including brute force, leaked seeds and malware.
- Multi-sig phishing attacks accounted for about 10% of losses and rose sharply quarter over quarter.
Context
Crypto losses in April 2026 were driven less by flaws in on-chain code and more by failures around access, signing, and operator trust. In practice, that means the critical control surface is no longer just the smart contract itself, but the identities and workflows that can approve transactions, move treasury funds, and change upgrade paths.
This is a classic governance problem with a strong identity angle. Keys, signer accounts, multi-sig participants, and privileged admin endpoints behave like high-risk non-human identities, but many organisations still govern them as if they were ordinary operational accounts. That gap turns phishing, leaked secrets, and weak approval processes into direct financial loss.
Key questions
Q: What breaks when signer trust is weaker than the multi-sig quorum suggests?
A: The approval model breaks when signers rely on the same interface, message context, or urgency cue, because quorum no longer means independent verification. In that state, a multi-sig can approve malicious calldata just as efficiently as a legitimate transfer. The control fails at the human decision layer, not the contract layer.
Q: Why do private key breaches create more damage than many smart contract bugs?
A: Private key breaches bypass the trust checks that smart contract audits are designed to assess. If the attacker controls the signing credential, the contract sees a valid instruction, not a malicious one. This makes the compromise more direct, more flexible, and harder to block once the key is exposed.
Q: What are the signs that transaction approval controls are failing?
A: Warning signs include blind signing, signers using the same front-end context, repeated urgent approval requests, inconsistent transaction decoding, and reliance on a single admin device or network. If the protocol cannot independently verify what each signer is approving, the approval path is already operating outside a safe boundary.
Q: Who should be accountable for treasury signing risk in crypto operations?
A: Accountability should sit jointly with the protocol owners, security leadership, and the operators who control signing workflows, because treasury access is a privileged business function. The governance standard should be that no single person, device, or UI path can authorise material movement without independent checks and documented ownership.
Technical breakdown
Why smart contracts were not the main failure mode
The article's core point is that the contracts mostly executed as written. That matters because a correct contract still cannot defend itself if an attacker has a valid signing path into treasury, upgrade, or withdrawal functions. In crypto operations, the contract often acts as the final enforcement point, while the real control decision happens earlier at key custody, signer verification, and transaction approval. Once an attacker controls a hot key or a signer's approval flow, the protocol faithfully processes the malicious action. Practical implication: move risk analysis upstream to the identities and devices that authorise transactions, not just the code that executes them.
Practical implication: treat signer identity and key custody as part of the control plane, not as separate operational detail.
How multi-sig becomes a process failure
A multi-sig only reduces risk when signers independently understand what they are approving. The article shows that blind signing, spoofed wallet interfaces, and social engineering let attackers turn a nominally distributed approval model into a single point of failure. If signers rely on the same front end, the same message context, or the same urgent business request, the quorum is administrative theatre rather than effective control. This is why multi-sig is not a security guarantee on its own. Practical implication: require independent transaction verification on separate trusted devices and assume UI compromise is part of the threat model.
Practical implication: design multi-sig around independent verification, not just numeric quorum.
Why operational infrastructure is now in scope
The article groups DNS hijacks, dependency abuse, misconfigured RPC endpoints, exposed admin keys, and compromised hot wallets into the same loss pattern. That is the right lens: the attacker does not need to break cryptography if they can reach the signing or delivery path that surrounds it. In effect, build pipelines, secret managers, wallet interfaces, and admin panels have become identity-bearing control points for value transfer. This broadens the attack surface from chain code to the entire transaction supply chain. Practical implication: apply the same review discipline to off-chain infrastructure that you already apply to smart contract logic.
Practical implication: extend security review to the full transaction supply chain, including front ends, DNS, CI/CD, and secret storage.
Threat narrative
Attacker objective: The attacker aims to obtain valid signing authority so they can move assets or change control paths without needing to exploit the underlying contract logic.
- Entry occurs through phishing, malware, leaked secrets, spoofed interfaces, or cloud and CI/CD exposure that reaches key material or signer workflows.
- Credential access follows when attackers obtain private keys, multi-sig approvals, or privileged admin credentials that can authorise treasury or bridge actions.
- Escalation happens when those compromised signing paths are used to drain funds, approve malicious calldata, or take over validator and bridge infrastructure.
- Impact is direct financial theft at scale, with protocol treasuries, exchanges, bridges, and individual wallets losing funds despite correct contract logic.
Breaches seen in the wild
- Secrets in Docker Hub images (RWTH Aachen study): A 2023 RWTH Aachen study found secrets in 8.5% of container images, and 275,269 internet hosts still using the leaked private keys.
Read and download The State of NHI & AI Agent Breach Report 2026, covering 200+ breaches impacting Non-Human Identities including AI Agents.
NHI Mgmt Group analysis
Operator risk is now the dominant loss vector in crypto. April 2026 shows that the industry has moved past the era where the main concern was contract correctness alone. When attackers can steal funds by compromising keys, signers, or approval workflows, the governing question becomes who can authorise value transfer and under what conditions. That is a governance problem as much as a technical one, and practitioners should treat signing authority as privileged access.
Private keys behave like high-value non-human identities. They are persistent, reusable, and capable of autonomous economic action once exposed. That makes them materially closer to machine identity risk than to ordinary user-account risk, especially when they control treasury, upgrade, or bridge functions. The control failure is not simply leakage, but the absence of lifecycle governance around issuance, custody, review, and revocation.
Multi-sig is a control pattern, not a control outcome. The article makes clear that quorum alone does not provide assurance if every signer trusts the same UI, device, or message context. A quorum can still approve the wrong transaction when independent verification is missing. For identity and access programmes, the lesson is to assess the quality of approval context, not just the number of approvals.
Crypto security is converging on the broader NHI problem space. The same control gaps seen here also appear in cloud and workload identity environments: exposed secrets, over-trusted service paths, and weak lifecycle management. The named concept here is signer trust collapse, meaning the point where transaction approval mechanisms stop providing meaningful assurance because the human and technical trust layers fail together. Practitioners should govern approval paths as privileged identity workflows, not as informal operations.
What this signals
Signer trust collapse: April's losses show that approval workflows fail when human verification, device trust, and transaction context are all assumed rather than enforced. For practitioners, the practical shift is to govern signing paths like privileged identities and to remove any single device or UI from being able to authorise material value transfer alone.
The broader lesson is that crypto teams are now facing the same governance problem seen in other identity-heavy environments: the real attack path often sits around the protected asset, not inside it. That makes key custody, approval context, and off-chain operational resilience the controls that need board-level visibility, not just contract audit results.
For practitioners
- Harden signer workflows Require independent transaction verification on separate trusted devices, and ban blind signing for any treasury, upgrade, or withdrawal action.
- Treat keys as privileged identities Inventory every private key, validator key, admin account, and signer role with the same ownership and lifecycle controls used for privileged access.
- Put time and spend limits on high-risk actions Use time-locks, approval thresholds, and spend caps on treasury transfers, bridge changes, and upgrade paths so a single compromise cannot empty value immediately.
- Extend review to the off-chain attack surface Review DNS, CI/CD, secret managers, wallet front ends, and admin panels as part of the same transaction risk chain, not as separate systems.
- Run signer drills and compromise scenarios Test what happens when a signer laptop, admin account, or hot wallet is compromised and measure whether the protocol can still contain loss.
Key takeaways
- April 2026's losses show that the most expensive failures in crypto now come from operator compromise, not from breaking smart-contract logic.
- Private keys, signers, and off-chain workflows are the real control surface when protocols can move value with a valid approval.
- Teams that still rely on contract audits alone are missing the governance layer where most losses now occur.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
OWASP Non-Human Identity Top 10 and MITRE ATT&CK address the attack and risk surface, while NIST SP 800-53 Rev 5 and NIST CSF 2.0 set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP Non-Human Identity Top 10 | NHI-02 — Secret Leakage | Leaked seeds, exposed keys, and misconfigured secret managers are a central loss path in the article. |
| NHI-05 — Overprivileged NHI | Treasury, bridge, and admin signing paths show excessive authority concentrated in a few keys. | |
| Recommendation — Scan and revoke exposed NHI secrets before they can be used to authorise treasury or signing actions. Reduce signing scope so no single NHI can authorise high-value transfers or control changes alone. | ||
| MITRE ATT&CK | TA0006;TA0008 — Credential Access; Lateral Movement | The article centres on credential theft followed by movement into signing and treasury functions. |
| Recommendation — Map key compromise and signer abuse to TA0006 and TA0008 to prioritise detection of privileged movement. | ||
| NIST SP 800-53 Rev 5 | IA-5 — Authenticator Management | The loss pattern is driven by poor credential lifecycle control for keys and signer secrets. |
| Recommendation — Apply IA-5 to govern issuance, rotation, storage, and revocation of private keys and signer credentials. | ||
| NIST CSF 2.0 | PR.AA-05 — Access Permissions, Entitlements and Authorizations | The article shows that authorisation scope, not just authentication, determines blast radius. |
| Recommendation — Use PR.AA-05 to validate who can sign what, and remove unnecessary authorisation paths from treasury workflows. | ||
Key terms
- Signer Trust Collapse: A failure mode where multiple approvers exist on paper, but each signer relies on weak interfaces or social cues instead of independently verifying the transaction. The control looks strong structurally, yet the human verification layer no longer distinguishes legitimate instructions from malicious ones.
- Privileged Signing Path: A privileged signing path is the set of people, devices, applications, and approvals that can authorise transfers, upgrades, or administrative changes. In practice, it behaves like privileged access control, because compromise of that path gives an attacker direct power over assets and governance actions.
- Blind signing: Approving an action without enough context to understand its real effect. In identity and transaction-heavy environments, blind signing turns human review into a weak control because the signer can be manipulated into authorising privileged actions they cannot reliably interpret.
- Operational Security Failure: An operational security failure is a mistake that exposes identity, intent, infrastructure, or movement patterns that should have remained separate. In cybercrime investigations, these failures often include reused accounts, overlapping logins, wallet reuse, or careless cross-channel behaviour that allows correlation across personas.
Deepen your knowledge
NHI Foundation Level course, the industry's only accredited NHI security programme, covers NHI governance, machine identity security, and secrets management. It helps security leaders translate identity controls into practical oversight for privileged workflows and high-risk credentials.
Published by the NHIMG editorial team on July 14, 2026.
Updated on October 13, 2026.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org