1

It is generally said that 2048-bit RSA keys are roughly equivalent to 112-bit symmetric keys and 2048-bit RSA keys are "sufficient for commercial use until around the year 2030" (https://security.stackexchange.com/a/65180/189448).

What is a rough estimate for when AES, RSA, and ECC keys of n bits are unsafe for amatuers?

"Unsafe" as in a tech-savvy person of medium income can decrypt a given message within 2 years.

  • One caveat I would add is that there's always the possibility of an exploit being found which takes the time to crack your key to a feasible time for an attacker. Like we saw today with the new TLS 1.3 vulnerability published. – Daisetsu Feb 10 at 3:58
  • @Daisetsu I'm sorry but I really don't see how the possible duplicate is a duplicate. Can you elaborate? – facjqhpcc Feb 10 at 4:02
  • @Daisetsu The TLS 1.3 vuln has been disclosed a few months ago. It's not really new. I'm also assuming the technology grows steadily and no sudden breakthroughs – facjqhpcc Feb 10 at 4:03
  • Oh wow. I don't know what happened to that link. Totally the wrong one. – Daisetsu Feb 10 at 4:03
  • Slight correction maybe: 2048-bit RSA keys are equivalent in strength to 112-bit symmetric keys not 128-bit. See here for source, via NIST I think. – Ellie Kesselman Feb 10 at 4:46
1

For this kind of purposes, there is a website, keylength.com, that uses various academic and private organizations mathematical formulas to approximate.

If you select 2030 you will get the results.

enter image description here

  • Yes, but these estimates are more for "safe against state actors" aren't they? I'm only looking for estimates against, say, a tech-saavy friend. – facjqhpcc Feb 23 at 7:13
  • It is about the resources you have. Have a look at this. – kelalaka Feb 23 at 7:16

Not the answer you're looking for? Browse other questions tagged or ask your own question.