Are Password Managers Safe (and Hackable) in 2026?
By Thomas Lovas Lokoy · Published September 6, 2026 — researched and drafted with AI assistance from published sources; edited and published by NorwegianSpark SA.
Yes, password managers can be hacked. One of the largest was, publicly and comprehensively. Anyone answering this question with a flat "no, they are perfectly safe" is either not paying attention or selling you something.
The useful question is different: given that they can be breached, is putting every password in one is still the better bet than the alternative? The answer is yes, and it is worth understanding exactly why, because the reasoning also tells you what to do to make it true for you specifically.
What actually happened to LastPass
This is the case everyone half-remembers, so here is what the company itself published rather than what the headlines said.
In its notice dated 22 December 2022, LastPass stated that the threat actor accessed backup data containing "basic customer account information and related metadata including company names, end-user names, billing addresses, email addresses, telephone numbers, and the IP addresses", and also obtained "a backup of customer vault data from the encrypted storage container".
On the encryption, LastPass stated that the sensitive fields "remain secured with 256-bit AES encryption and can only be decrypted with a unique encryption key derived from each user's master password using our Zero Knowledge architecture", and that "the master password is never known to LastPass and is not stored or maintained by LastPass". It added that "the encryption and decryption of data is performed only on the local LastPass client".
Read that carefully, because it contains both the reassurance and the warning.
The reassurance: attackers took the vaults and could not simply open them. The architecture did the job it was designed for. Compare that to a breach of an ordinary service, where the stolen data is immediately readable.
The warning: they took the vaults. Encrypted data that is in someone else's hands is data they can attack offline, forever, at their own pace, with as much computing power as they care to buy. There is no rate limit and no lockout on a file sitting on a criminal's disk. The only thing standing between that file and its contents is the strength of the master password and the difficulty of the key derivation.
That is the honest shape of the risk, and it is why "zero-knowledge" is a genuine protection and not a magic word.
The three ways a vault actually falls over
| Failure | What it looks like | What stops it |
|---|---|---|
| Weak master password | A stolen vault is cracked offline, months or years later | A long, unique master password nothing else uses |
| Malware on your device | A keylogger or infostealer captures the master password as you type it, and the vault opens normally | Endpoint protection, updates, not running untrusted software |
| No second factor | A leaked or guessed master password is sufficient on its own | Two-factor on the manager itself, ideally hardware or app-based |
Notice what is not on that list: "the vendor's encryption gets broken". That is not the realistic failure mode and it never has been. Every real-world compromise of a password manager runs through one of those three rows, and two of the three are on your side of the line rather than the vendor's.
The middle row deserves particular attention because it defeats every architecture. If malware is running on your machine with your privileges, it can watch you type the master password and then read the decrypted vault out of memory. No amount of AES protects against a keylogger. This is the reason a password manager and endpoint protection are complementary rather than alternatives — see how your data ends up on the dark web for how infostealers work in practice.
What the standard actually says
It is worth knowing that using a password manager is not merely tolerated by security guidance — it is written into it.
NIST Special Publication 800-63B, section 3.1.1.2, states that "Verifiers SHALL allow the use of password managers and autofill functionality", and that verifiers "SHOULD permit claimants to use the 'paste' function when entering a password to facilitate password manager use when password autofill APIs are unavailable". "SHALL" is the strongest normative term in that document. A site that blocks paste in its password field is not being careful; it is out of step with the standard.
The same section requires a minimum of 15 characters for a password used as a single factor, tells verifiers to support at least 64, and states that they "SHALL NOT impose other composition rules (e.g., requiring mixtures of different character types)" and "SHALL NOT require subscribers to change passwords periodically". Our separate guide covers the password rules NIST says to stop using in full — this page cites the standard only to answer the safety objection, not to re-tread that ground.
The alternative you are actually comparing against
The question is never "password manager or perfect security". It is "password manager or what I am doing now", and what most people are doing now is one of these:
- The same password everywhere. One breach at any site becomes access to all of them. This is the arrangement password managers exist to end, and it is still the most common one.
- A memorable pattern with variations. Attackers know the patterns. A password reused with a site name appended is not a set of distinct passwords.
- Saved in the browser. Better than nothing and genuinely improving, but it is the exact store that infostealer malware is written to empty, and it usually travels less well across devices and browsers.
- A notebook by the desk. Immune to remote attack and a real option for the accounts you rarely touch, if you are honest about the physical risk and about whether you will keep it up to date.
Against every one of those, a dedicated manager wins on the metric that matters: how many accounts fall when one site is breached. With a manager, the answer is one. Without, it is all of them.
Making yours actually safe
- Make the master password long and unique. Length beats complexity. A passphrase of several unrelated words is easier to remember and harder to crack than a short string of symbols. It must not be a password you use anywhere else, ever.
- Turn on two-factor authentication on the manager itself. This is the single highest-value setting in the product and it is frequently left off.
- Store the recovery kit somewhere offline. A manager you cannot get back into is its own disaster, and a recovery code emailed to yourself defeats the point.
- Keep endpoint protection running. The keylogger row above is the one you cannot architect around.
- Prefer passkeys where they exist. A passkey has no shared secret to steal or phish, and most good managers now store them alongside passwords.
- Use the built-in breach check. Most managers can tell you which stored passwords have appeared in a known breach, which turns the vault into a to-do list rather than a filing cabinet.
Which product matters far less than whether you use one. If you want the comparison, it is in best password managers, with individual writeups for Proton Pass and NordPass; both are available directly as Proton Pass and NordPass. The household question — sharing logins without emailing them — is in sharing passwords safely.
The counter-argument, taken seriously
The strongest objection is the one about concentration: a password manager creates a single point of failure, and single points of failure are exactly what security engineering tries to avoid. That objection is correct as far as it goes.
What it misses is that the alternative is also a single point of failure — it is just distributed and invisible. If you reuse one password across forty sites, then all forty of those sites are your single point of failure, and you do not control the security of any of them. A manager moves the concentration to a place you can defend: one long master password, one second factor, one machine you keep clean. That is a smaller and far more defensible surface than forty companies of unknown competence.
The second objection, about trusting a vendor, is answered by the architecture rather than by the vendor's promises. In a zero-knowledge design the provider cannot read your vault even if it wanted to — which is exactly what the LastPass notice describes, and exactly why the stolen vaults were not simply published.
Yes, they are hackable. Use one anyway, make the master password long, turn on the second factor, and keep the device clean. Which one to use is a separate question, and our security tool comparison answers it. This is general guidance, not security advice for a specific threat model.
Affiliate disclosure
This article contains affiliate links. If you purchase through them, CyberTechVault earns a commission at no extra cost to you. Our assessments are based on vendors' published documentation, independent lab results and security disclosures — not on hands-on testing by us. Affiliate relationships never decide what we recommend.
Full disclosure: /affiliate-disclosure.
Sources
Factual claims above were checked against these primary sources. Verify directly on the source for anything time-sensitive before relying on it.
Continue reading
privacy
Black Friday Scams 2026: Spot a Fake Deal Before You Pay
Black Friday falls on 27 November this year. The scams that peak around it are predictable: fake shops, too-good ads, odd payment demands and delivery texts. Here is how to check a deal in two minutes.
antivirus
How to Remove Malware from Mac and Get Rid of Adware in 2026
Mac infections are overwhelmingly adware and browser hijacks, not viruses. The removal order that works, using Apple silicon and Intel safe mode correctly.
guides
Best Identity Theft Protection Services 2026: What Works
The best identity theft protection in 2026 is a layered strategy, not one product. The layers that matter and the tools we recommend.
