Posted by Martijn Grooten on Dec 23, 2019
At the end of this month, I will step down as Editor of Virus Bulletin. Before doing so, I have been sharing some 'parting thoughts' in five blog posts, based on my experience working in the IT security industry. This is the final post in the series, the previous ones were: 'cybersecurity as a social science', 'the need for education in security', 'taking security seriously' and 'the big picture'.
IT security is about things that break, and thus a lot of security news is bad news. Not all news is bad though: sometimes botnets are taken down, cybercriminals get arrested and security technologies such as HTTPS see ever wider adoption. There is also, however, a more subtle kind of good security news that we often fail to see.
One example is Android malware. As anyone working in security will tell you, there is a lot of it. And a lot of Android malware is found on the official Google Play store, sometimes exceeding a million downloads.
But if you look deeper, you will notice that most of that malware performs relatively harmless activities such as click-fraud or showing unwanted ads. Experts may argue that it could have done worse things, but it doesn't, and that is crucial. Stealing private data, for example, is quite hard for Android malware to do because of the Android security model and would be a lot easier to detect.
To some extent, the same applies to botnets that mine for cryptocurrencies on Windows machines: the fact that they don't engage in more harmful activities should be telling. Here, something else plays a role too: many of these botnets reside on older, often unpatched or underpatched devices typically in lower-income countries. Large-sized Windows botnets in high-income countries may be a thing of the past and, maybe in the not too distant future this will be the case globally too.
Vulnerabilities often make the news, and here too the news is often good in a subtle way: 'zero-click exploits' have become quite rare. Most vulnerabilities merely break the security model, but still require a special condition to be met (such as code execution, a man-in-the-middle position or a user opening a link).
Earlier this year I gave a talk on 'Mitigation', the main point of which was that we are really bad at stopping attacks, but surprisingly good at mitigating them. We don't give ourselves enough credit for this and I believe it would help our understanding of the threat landscape if we did.