If you follow the links, you'll end up here:
<a href="https://www.usenix.org/conference/usenixsecurity16/technical-sessions/presentation/svenda" rel="nofollow">https://www.usenix.org/conference/usenixsecurity16/technical...</a><p><i>Can bits of an RSA public key leak information about design and implementation choices such as the prime generation algorithm? We analysed over 60 million freshly generated key pairs from 22 open- and closedsource libraries and from 16 different smartcards, revealing significant leakage. The bias introduced by different choices is sufficiently large to classify a probable library or smartcard with high accuracy based only on the values of public keys. Such a classification can be used to decrease the anonymity set of users of anonymous mailers or operators of linked Tor hidden services, to quickly detect keys from the same vulnerable library or to verify a claim of use of secure hardware by a remote party.</i>
>To help solve this serious issue, Enigma Bridge is proud to have developed a cost-effective, ground-breaking hardware security service which is based in the cloud.<p>Uh huh.<p>So <a href="https://dan.enigmabridge.com/re-investigating-the-origins-of-rsa-public-keys/" rel="nofollow">https://dan.enigmabridge.com/re-investigating-the-origins-of...</a> has some more details, and the paper is at <a href="https://www.usenix.org/conference/usenixsecurity16/technical-sessions/presentation/svenda" rel="nofollow">https://www.usenix.org/conference/usenixsecurity16/technical...</a><p>It took me a couple of reads through the article to work out they're not necessarily talking about key strength, but fingerprinting the software / hardware that created the key in the first place.
The tool this seems to be based on is here: <a href="http://www.fi.muni.cz/~xsekan/" rel="nofollow">http://www.fi.muni.cz/~xsekan/</a>
In other news, it is less likely that an SSL key was generated using IIS if the platform it is running on is AIX.<p>I mean, it's a useful tip when targeting a relatively dark target, but at the same time it isn't an absolute indicator of anything other than what generated the key (which nobody was really surprised at before, since everything from the prng to implementation differences could result in unique signatures for keys).<p>This will be useful when someone finds an implementation-specific hole in a key gen and someone wants to sweep the internet for servers with bad keys.