Regin, advanced malware that is most likely a government espionage tool, is making headlines.
This is because it's a very well-constructed set of tools and also because observers are surprised at how successful it was. It also targeted GSM networks, which is novel.
The big question is, how could the major anti-malware firms have missed this threat for so long?
Or, one might ask, did they really miss it or quietly detect it?
Some people appear to believe that, as Regin was probably created and used by Western governments, then Western anti-malware companies colluded to ignore the threat.
Symantec seems to have been slow to notice Regin because its write-up of Backdoor.Regin claims that it was discovered in December 2013, which is much later than March 2011, when Microsoft updated its definitions to include Regin.A.
In an effort to find a history of Symantec's detection of this malware I obtained an archive of Regin samples from security researcher Claudio Guarnieri and asked the kind folk at VirusTotal to discover when, if ever, Symantec's scanner first detected each sample.
Before we look at these results I want to be clear about what these results mean and what they do not because VirusTotal data is easily abused and dodgy conclusions readily-reached.
The table below indicates that Symantec's technology was capable of detecting most of the samples as being at least suspicious from February 2010. It then made a clearer classification of being a 'Trojan' from March 2011.
Only yesterday (24th November 2014) did it officially label the threat as 'Regin'. This corresponded to its announcement of the Regin threat.
Usually the problem with using VirusTotal is that someone will upload some files, show that product X failed to recognise them and then conclude that the product, or the entire anti-virus industry, is useless.
In this case we can see dates relating to when the product detected the files as threats. Possibly the product would have protected against these files even earlier, and possibly those that appear as having been missed (Classification = 'nothing') would have been stopped through some other layer of protection not related to file signatures.
So I see the following as a worst-case scenario. Symantec's scanner recognised most of these files as threats from around 2011 onwards. Maybe it was capable of stopping them and maybe not - we can't know that for sure. But it's fair to assume that if a signature-based scanner can recognise a file then it will probably generate an alert at the very least.
I've focussed one Symantec simply because it first announced the Regin malware, minutes before other vendors joined in.