The CyberSignal
  • Latest
  • Trending
  • Cyber Attacks
  • Data Breaches
  • Threat Intelligence
  • Critical Infrastructure
  • Policy & Government
  • Cybersecurity 101
  • Vulnerabilities
  • About Us
  • Weekly Briefing
Account Takeover (ATO)

Dashlane Says Hackers Beat Its 2FA and Downloaded About 20 Customers' Encrypted Vaults

Dashlane now says the brute-force attack it disclosed on May 31 succeeded: by defeating 2FA on about 20 customer accounts, attackers downloaded copies of those users' encrypted password vaults. The vaults stay locked behind each user's master password, but affected users should rotate.

Nicholas Robert

Nicholas Robert

03 Jun 2026 — 8 min read
Share
Line-art of a vault door ajar with a small key icon being carried out on a thin line; the vault's combination dial shows a single flat red dot.

Key Takeaways

  • Dashlane disclosed on June 2, 2026 that the brute-force attack it first reported on May 31 succeeded against about 20 customer accounts, and that the attackers downloaded copies of those users' encrypted password vaults.
  • The attack brute-forced Dashlane's two-factor authentication to register new devices on existing accounts — automated software rapidly guessing every numeric combination before the short-lived 2FA code expired; Dashlane found no evidence its own systems were compromised and has not said how 2FA was defeated.
  • The stolen vaults remain encrypted with each user's master password, which Dashlane never receives in plaintext, so decryption hinges on master-password strength; affected users should rotate their master password and treat saved credentials as at-risk, prioritizing rotation by sensitivity.

Yesterday the story was that Dashlane's lockout fired. Today the story is that it didn't catch everything: on a small set of accounts the brute-force won, and the encrypted vaults left the building.

NEW YORK — Dashlane disclosed on June 2, 2026 that the brute-force attack it first reported on May 31 succeeded against about 20 customer accounts, and that the attackers downloaded copies of those users' encrypted password vaults after defeating the company's two-factor authentication.

It is an escalation from the original disclosure, which The CyberSignal covered as a triggered-lockout event: the news then was that Dashlane's protections fired; the news now is that on a small number of accounts they did not catch everything, and vault data left for attacker-controlled systems.

Incident Overview
FieldDetails
VendorDashlane (password manager)
DisclosureJune 2, 2026 update to the brute-force advisory Dashlane first posted around May 31
ScopeAbout 20 customer accounts (Dashlane: 'the 20 or so customers'); reporting describes them as personal-subscription users
MechanismBrute-force against two-factor authentication to register new devices on existing accounts — automated guessing of every numeric combination before the short-lived 2FA code expired
ResultAttackers downloaded copies of the affected users' encrypted password vaults
Dashlane systemsNo evidence Dashlane's own systems were compromised; the company has not said how the 2FA protection was defeated
Vault protectionVaults remain encrypted with each user's master password, which Dashlane says is never uploaded in plaintext; users with easily guessed master passwords are at greater risk of decryption
AttributionUnknown — Dashlane has not said who was behind it, whether specific customers were targeted, or whether any ransom demand was made
ResponseDashlane notified the affected users and says it has 'taken steps to mitigate the risk of future incidents' (unspecified)

What Happened

In an update to the security advisory it first posted around May 31, Dashlane said on June 2, 2026 that attackers brute-forced its two-factor authentication system and gained access to about 20 customer accounts — and that, having defeated 2FA, they downloaded copies of those customers' encrypted vaults, which store passwords and other sensitive credentials. In Dashlane's words, 'the goal of the attack was to brute-force two-factor authentication (2FA) protections to allow the attacker to register new devices on existing user accounts,' using automated software to 'rapidly submit every possible numeric combination to the system, hoping to guess the exact sequence before the short-lived security code expires.' The company said it has notified the roughly 20 affected customers.

Two qualifiers matter and Dashlane states both. First, the company says there is no evidence its own systems were compromised, and it has not yet explained how the attackers defeated its two-factor protections. Second, the stolen vaults are scrambled and cannot be read without the customer's master password, which Dashlane says is known only to the customer and is never uploaded to Dashlane in plaintext. The catch Dashlane itself flags is that customers with an easily guessed master password are at greater risk of having it cracked and their vaults decrypted — the same dynamic that turned the 2022 LastPass vault theft into years of follow-on credential and cryptocurrency theft against users whose master passwords were weak.

From 'Lockout Triggered' to 'Vaults Downloaded'

The escalation is the story. The CyberSignal covered Dashlane's original disclosure as a brute-force attack that targeted the new-device token flow and triggered account lockouts — the framing there was that the protection worked, locking accounts as designed. The June 2 update revises that picture: on about 20 accounts the brute-force succeeded before the lockout fully contained it, the attackers registered as a new device, and they pulled down the encrypted vault. That is a categorical difference. A locked-out account is an attempted intrusion; a downloaded vault is exfiltrated data sitting on attacker infrastructure, where it can be attacked offline at the attacker's leisure with no rate limit and no lockout to stop them.

Why an Encrypted Vault Is Still a Real Problem

It is tempting to file 'the vaults are encrypted' as reassurance, and it is genuinely the thing standing between the affected users and total compromise. But encryption converts the problem from an immediate breach into a race. Once a vault is in the attacker's hands, the only barrier is the user's master password, and the attacker can run an offline brute-force or dictionary attack against it indefinitely. A long, random, unique master password may never fall; a short or dictionary-word master password could be cracked in hours to days. That is precisely the gradient Dashlane points to when it warns that customers with easily guessed master passwords are at greater risk — and it is why the responsible posture for an affected user is to assume the vault contents may become readable within the brute-force window and to act before that window closes, rather than to treat encryption as a guarantee.

A Recovery-Construct Attack, in Familiar Company

Strip away the password-manager specifics and this is another attack on the machinery that lets a user prove who they are — the recovery and re-authentication layer that The CyberSignal has watched come under sustained pressure this cycle. It sits with the Signal recovery-key phishing wave, the Meta AI support bot tricked into account takeovers, the Tycoon2FA kit that defeats multi-factor authentication, and the Microsoft Android token-isolation exposure. Each one targets the moment of identity verification rather than a stored password, and together they track the trend Verizon's 2026 DBIR captured of attackers concentrating on the access path. A password manager is the highest-value single instance of that target, because one cracked master password unlocks everything behind it.

Scope and Impact

The confirmed scope is small and specific: about 20 personal-subscription accounts, each of which had its encrypted vault downloaded, and all of which Dashlane says it has notified. There is no indication this was a mass event affecting the broader user base, and Dashlane reports no compromise of its own infrastructure. What is not yet public is consequential, though: Dashlane has not explained how its 2FA was defeated, has not said whether the ~20 customers were singled out for who they are, and — per the reporting — enterprise Business and Teams tiers have not been explicitly addressed, so organizations using Dashlane at work should confirm their tier's status with the vendor rather than assume the personal-plan disclosure covers them.

The blast radius for an affected individual, by contrast, is potentially their entire digital life. A password-vault compromise is unique among breaches because the vault is, by design, the index of every account the user has. If the master password falls, the attacker inherits banking, email, work, healthcare and every other saved credential at once. That is why the operational response cannot be scoped to Dashlane itself; it has to extend to every service whose credentials the user stored, which is the practical definition of a Tier 1 personal-security incident. The historical precedent is instructive and sobering: in the 2022 LastPass case, the encrypted vaults that were stolen sat dormant for months before researchers began tracing large-scale cryptocurrency thefts back to master passwords cracked offline — a reminder that the damage from a stolen vault can surface long after the initial disclosure, and that 'no fraud yet' is not the same as 'no exposure.'

Response and Attribution

For any user Dashlane has notified, the immediate runbook is: rotate the Dashlane master password now, which protects against future decryption attempts even if the old vault copy is cracked; then treat every credential saved in the vault as at-risk and rotate it, prioritizing by sensitivity — financial accounts, primary email, work identity and healthcare first, lower-stakes services next — and re-enable two-factor authentication everywhere, preferably with an authenticator app or passkey rather than SMS. Monitor financial and email accounts for unauthorized access for at least 90 days, since an offline master-password crack may take weeks; and watch for phishing that impersonates Dashlane's breach notification, a predictable follow-on whenever a vendor emails affected users about an incident. Users who were not notified do not need to panic, but rotating to a long, random master password and pruning unrecognized devices in the Dashlane app is cheap insurance given that the attack proved 2FA-protected vaults can be reached with enough brute-force budget.

For product-security leaders, the transferable lesson is about the enrollment flow, not Dashlane specifically. New-device-registration and 2FA-challenge endpoints are now a documented brute-force target, and even a vendor with mature auto-protection had an edge case where guessing succeeded. Audit your own product's device-enrollment and 2FA flows for strict rate-limiting, exponential backoff, and per-account lockout thresholds that cannot be outrun by automated guessing against a short-lived numeric code — and where feasible move enrollment to phishing-resistant, FIDO2/passkey-based mechanisms that eliminate the brute-forceable numeric-code class entirely. On attribution, the honest position is Dashlane's: the actor, the motive, and any ransom dimension are unknown, and it would be speculation to assign them.


The CyberSignal Analysis

Signal 01 — A Downloaded Vault Changes the Clock, Not Just the Status

The difference between yesterday's story and today's is not cosmetic. An attempted login that gets locked out is a contained event; a downloaded encrypted vault is a deadline. Offline, the attacker faces no rate limit, no lockout, and no expiry — just the user's master password and however long it takes to guess it. That reframes the defender's job from 'was the intrusion stopped?' to 'how strong is the master password, and can we rotate everything it protects before it's cracked?' For users, the takeaway is that encryption buys time, not safety, and time should be spent rotating. For the industry, it is a reminder that the moment data leaves your perimeter, your controls stop applying and the math shifts entirely to the strength of the one secret you never held.

Signal 02 — The Master Password Is the Whole Ballgame

Dashlane's architecture did exactly what zero-knowledge design is supposed to do: the vaults left encrypted, and Dashlane never held the keys. That is the reason this is a serious incident rather than a catastrophic one. But it also concentrates the entire residual risk onto a single user-chosen secret, and the 2022 LastPass aftermath showed what happens when that secret is weak — years of follow-on theft against the minority of users whose master passwords could be brute-forced. The lesson for every password-manager user, not just Dashlane's, is that the master password is not just one password among many; it is the single point of failure for the entire vault, and it deserves length and randomness commensurate with that role. A passphrase of real entropy is the difference between a stolen vault that is a non-event and one that is a slow-motion disaster.

Signal 03 — Enrollment Flows Are the New Brute-Force Frontier

The technical heart of this attack was not a stolen password or a software vulnerability; it was guessing a short-lived numeric 2FA code fast enough to register a new device. That is an enrollment-flow weakness, and it generalizes far beyond Dashlane. Any service that gates new-device registration or account recovery behind a short numeric code is exposed to the same automated-guessing pressure, and the defenses are well understood but unevenly applied: strict and escalating rate limits, hard per-account lockouts, anomaly detection on enrollment attempts, and — the durable fix — replacing guessable numeric codes with phishing-resistant passkeys. The CyberSignal's recurring observation this cycle is that attackers have moved to the identity-verification layer; this incident is the password-manager industry's instance of that shift, and the mitigation is to harden enrollment as rigorously as authentication.


Sources

TypeSource
OfficialDashlane — Security Advisory: Brute-Force Attack on Dashlane User Accounts
ReportingTechCrunch — Password Manager Dashlane Says Hackers Stole Some Customers' Password Vaults
ReportingSecurityWeek — Dashlane Brute-Force Attack Leads to Limited Encrypted Vault Downloads
The Hacker News — Dashlane Discloses Brute-Force Attack on User Accounts

Read more

Editorial line-art illustration showing a shield, bug, calendar, lock, and ranked vulnerability list with a red marker, representing the top CVEs and cyber threats of May 2026.

The Top CVEs of May 2026: Edge Devices Under Active Attack While Patch Tuesday Goes Quiet

Microsoft shipped its first zero-day-free Patch Tuesday since June 2024 — but the month's real action was elsewhere: a CISA Emergency Directive for Cisco SD-WAN, exploited PAN-OS flaws, and a Drupal core SQL-injection, all under active attack.

04 Jun 2026
Editorial science-poster illustration of incident response planning symbols — a binder, a phone, a checklist, a stopwatch, a fire extinguisher, and a shield.

What Is an Incident Response Plan?

A clear guide to incident response plans — what they are, why every organization needs one, what they should contain, and how to build, test, and maintain one.

03 Jun 2026
Line-art of a planning table joining a cyber-network node, a shield and a small AI chip into one operations map; the chip carries a flat red dot.

Pentagon's Top Cyber Official Wants Cyber in Every Operation — and Security Built Into AI From Day One

The Pentagon's top cyber official, Katherine Sutton, says the Defense Department must pull cyber 'out of its silo' and build it into every operation from day one — and must bake security into the AI tools it adopts, rather than treating it as an afterthought.

03 Jun 2026
Line-art of an executive-order document with a seal linked to a government building and utility icons by thin lines; the seal carries a flat red dot.

Trump Signs a Scaled-Back AI Executive Order Built Around Sharing AI-Found Vulnerabilities With Critical Infrastructure

Trump signed an executive order on June 2 setting up a voluntary framework for the government to vet 'covered frontier' AI models for up to 30 days before release and to share AI-found vulnerabilities with critical-infrastructure operators — notably narrower than an earlier draft.

03 Jun 2026
The CyberSignal
  • Daily Briefing
  • Weekly Briefing
  • Corrections
  • Privacy Policy
Powered by Ghost