Common Password Myths, Debunked
Some of the most repeated password advice is outdated, oversimplified, or was never quite right to begin with. Here are the myths that persist anyway, why they took hold in the first place, and what's actually true according to current guidance.
Myth: You should change your password every 30–90 days
Modern guidance from NIST and other standards bodies actively recommends against mandatory periodic rotation for accounts without evidence of compromise. Forced rotation tends to produce small, predictable variations of the same password rather than genuinely new ones — see our password rotation guide for the full reasoning behind this shift.
This myth persists mainly because it was standard corporate policy for so long that it became assumed wisdom, even after the research behind it was revisited and the guidance changed. Many organizations are still catching up to the updated recommendation.
Myth: Adding a symbol or number automatically makes a password strong
Swapping "e" for "3" or appending "!" is such a common pattern that cracking tools test for it by default. It adds far less protection than it feels like it should. Length and genuine randomness matter more than hitting a character-type checklist — see our entropy guide for the underlying math.
Myth: Security questions are a solid backup layer
Answers to questions like "mother's maiden name" or "first pet" are often discoverable through public records or social media, or guessable through common answers. Treat them as a weak point, not a security feature — some services now let you enter a random string instead of a real answer, which is worth doing where available.
Myth: A password manager is a single point of failure to avoid
The alternative — memorizing or reusing passwords — creates far more actual risk. Reputable password managers use encryption specifically designed so the company itself can't read your stored data, meaning even a breach of their servers doesn't directly expose your passwords. See our password managers explained guide for the mechanics behind this.
Myth: Longer passwords are inconvenient without real benefit
The inconvenience is real for passwords you type manually, but a password manager eliminates that cost by autofilling — so length stops being a tradeoff once you're not the one typing it every time.
Myth: My accounts aren't valuable enough to be targeted
Most attacks aren't personally targeted at all — they're automated, testing millions of leaked credentials against every major service simultaneously. Your account doesn't need to be individually interesting to end up in that sweep; it just needs a password that was reused somewhere that got breached.
Myth: A password with a mix of upper and lowercase is automatically random
Capitalizing the first letter of a word — the single most common capitalization pattern humans use — barely changes the effective guessing difficulty, because cracking tools already test the capitalized version of every dictionary word by default. True randomness means the capitalization itself is unpredictable, not just present.
Myth: If a site doesn't require a strong password, a weak one is fine
A site's password requirements reflect its own minimum bar, not a recommendation. Plenty of sites still allow surprisingly weak passwords for compatibility or legacy reasons — that doesn't mean a weak password is actually safe there, only that the site won't stop you from choosing one.
Myth: Antivirus software makes weak passwords less risky
Antivirus software protects against malware running on your device — it does nothing to stop a leaked or guessed password from being used directly against a website's login page from somewhere else entirely. These are separate layers of protection addressing separate threats, and one doesn't substitute for the other.
Myth: Using your browser's autofill is inherently insecure
Browser autofill is generally safe for the passwords it stores, and it does at least discourage manual reuse. It's simply a less feature-rich option than a dedicated password manager — missing things like breach monitoring, secure sharing, or cross-browser sync — not a security risk in itself.
Frequently asked questions
If I shouldn't rotate passwords regularly, when should I change one?
Change a password immediately if there's evidence it was exposed — a breach notification, a failed login attempt you didn't make, or a positive result from a breach checker. Otherwise, a strong, unique, unexposed password can stay in use indefinitely.
Are security questions ever safe to use?
Only if you treat them as a second password rather than a real answer — enter a random, unrelated string instead of your actual mother's maiden name, and store that string in your password manager like any other credential.
Is it true that password managers get hacked more than individual accounts?
No good evidence supports that framing. The relevant comparison isn't "manager vs no manager" but "unique generated passwords protected by strong encryption" vs "reused, memorized, weaker passwords" — the former is safer in essentially every realistic scenario.
Does using a passphrase instead of a random password count as a 'weaker' choice?
No, as long as the words are genuinely randomly selected rather than a memorable phrase you already knew. A random passphrase can reach the same entropy as a random character string, just with better memorability.
Why do these myths persist if they're outdated?
Much of this advice was genuinely reasonable guidance years ago, or was simplified for easy communication, and it's had a long time to spread through IT policies, onboarding documents, and word of mouth. Guidance updates faster than institutional habits do.
Is it a myth that longer passwords are always better?
Not quite a myth, but it has a practical limit — once a password is fully random and reaches a comfortable length (16-20+ characters), additional length adds security that's already far beyond what any realistic attack could exhaust.
Is there any truth to the idea that hackers 'target' specific people?
Targeted attacks do happen, particularly against high-profile individuals, but they're a small minority of overall attacks. The vast majority are automated and untargeted, which is actually good news: basic hygiene defeats most of what's actually out there.
Conclusion
Most password myths trace back to advice that made sense in an earlier context, or that got oversimplified along the way. The current, evidence-based version is simpler than the myths suggest: long, random, unique passwords, stored in a manager, backed by two-factor authentication where available.
Related articles
Common Password Mistakes
Ten everyday habits that quietly put your accounts at risk, with a clear fix for each one.
Read article →Password Length vs Complexity
The math behind entropy explains why a long password beats a short, complicated one.
Read article →Should You Rotate Your Passwords?
Forced password rotation used to be standard advice. Current guidance says otherwise — here's the reasoning, and when changing a password st
Read article →NIST Password Guidelines Explained
NIST's password guidelines reshaped how the industry thinks about password policy. Here's what SP 800-63B actually recommends, in plain lang
Read article →Free tools for this guide
Password Generator
Create a strong, random password in one click, right in your browser.
Open tool →Password Strength Checker
See entropy, crack-time estimates, and tips for any password you type.
Open tool →Passphrase Generator
Build a memorable Diceware-style passphrase with real entropy behind it.
Open tool →Username Generator
Memorable, random usernames for new accounts, games, and forums.
Open tool →PIN Generator
Random numeric PINs with optional repeating and sequential-digit avoidance.
Open tool →Random String Generator
Fully random strings for API keys, tokens, coupon codes, and test data.
Open tool →Hash Generator
Generate MD5, SHA-1, SHA-256, SHA-384, and SHA-512 hashes instantly.
Open tool →Base64 Encoder / Decoder
Convert text to and from Base64 instantly, with full Unicode support.
Open tool →UUID Generator
Generate RFC 4122 v4 UUIDs, one at a time or in a batch.
Open tool →