Most security advice is about protecting the login. This door is about the data that was already captured and is simply waiting.
Confirmed
Security researchers describe "harvest now, decrypt later" (HNDL) as an active, ongoing strategy: adversaries capture encrypted traffic today specifically to decrypt it once a cryptographically relevant quantum computer exists. Data protected by today's public-key schemes such as RSA or elliptic-curve key exchange is considered exposed to this risk. Separately, Palo Alto Networks' Unit 42 2026 Global Incident Response Report, based on more than 750 major incidents across more than 50 countries, found that in the fastest cases attackers went from initial access to data exfiltration in 72 minutes, about four times faster than the year before. Identity weaknesses played a material role in nearly 90% of the investigations, and attackers increasingly use AI for reconnaissance, phishing and scripting. Faster capture matters for HNDL because capture is the stage that counts.
Genuinely uncertain
When a quantum computer capable of actually breaking that captured data arrives. Estimates are commonly placed around 2030, but experts diverge meaningfully on the pace, and the sources we reviewed offer no firm date. The deadlines that do exist are policy deadlines, not predictions. In its 2024 draft guidance (NIST IR 8547, not yet final when we checked) NIST proposed deprecating quantum-vulnerable public-key algorithms for federal systems by 2030 and disallowing them by 2035. Those dates describe when organizations must have moved, not when the threat arrives.
Worth examining honestly
Your password was never really the target. A password protects a login. HNDL doesn't care about your login. It cares about the encrypted session itself, sitting in an archive, waiting. Changing a password after a breach fixes access. It does nothing for data that was captured in transit years earlier and is simply waiting for the math to catch up.
CIPHER-7's warning has always been about this exact distortion: the security theater of "change your password" while the actual exposure, encrypted archives collected over a decade and sitting in someone else's storage, goes unaddressed because it's not a login problem. It's a cryptography problem, and it's retroactive. The sensible way to think about it is shelf life. If a secret needs to stay secret for 15 years, and a break could plausibly arrive within 10, then it is already at risk today. A shopping list is not in that category. Medical records, legal files, trade secrets and diplomatic cables are.
The Door This Opens
The uncomfortable math: any sensitive data you sent under standard encryption more than a few years ago, that still needs to stay confidential today, may already be sitting in an archive somewhere. The fix exists. NIST finalized its first post-quantum cryptography standards in August 2024: FIPS 203 (ML-KEM, for key exchange), FIPS 204 (ML-DSA, for digital signatures) and FIPS 205 (SLH-DSA, a hash-based signature alternative). What doesn't exist yet, for most organizations, is the migration.
That migration is hard in ways that have nothing to do with the math. The new algorithms use larger keys and messages, which has caused compatibility problems with some load balancers, firewalls and devices that assume traditional size limits. Embedded and industrial equipment often has its cryptography fixed at manufacture and may need replacing, not patching. And no single scan finds every place cryptography is used, so even the inventory is incomplete. For an individual, the practical steps are modest: keep software updated, prefer services that announce post-quantum support, and think twice about what you store long term. For an organization, the first step is a cryptographic inventory, because you cannot migrate what you cannot find.