Slashdot Mirror
Exodus Intelligence Details Zero-Day Vulnerabilities In Tails OS
New submitter I Ate A Candle (3762149) writes Tails OS, the Tor-reliant privacy-focused operating system made famous by Edward Snowden, contains a number of zero-day vulnerabilities that could be used to take control of the OS and execute code remotely. At least that's according to zero-day exploit seller Exodus Intelligence, which counts DARPA amongst its customer base. The company plans to tell the Tails team about the issues "in due time", said Aaron Portnoy, co-founder and vice president of Exodus, but it isn't giving any information on a disclosure timeline. This means users of Tails are in danger of being de-anonymised. Even version 1.1, which hit public release today (22 July 2014), is affected. Snowden famously used Tails to manage the NSA files. The OS can be held on a USB stick and leaves no trace once removed from the drive. It uses the Tor network to avoid identification of the user, but such protections may be undone by the zero-day exploits Exodus holds.
The company plans to tell the Tails team about the issues "in due time"
I'm 100% certain "in due time" would come a lot sooner if the Tails OS maintainers coughed up the right fee, which means that this is most definitely NOT responsible disclosure.
I get that security researchers have to eat too, but damn - this sort of reeks of extortion. Maybe I'm wrong, but I know if I had a code project and some company said they knew I had holes but refused to tell me upon asking, extortion would be the first effing thought that would come to mind.
Quo usque tandem abutere, Nimbus, patientia nostra?
So they are selling vulnerabilities to hackers rather than telling the source maintainers? That's irresponsible at best.
This sounds like FUD against Tails. A security research firm finds some undisclosed zero-days in Tails, but doesn't describe what they could do - arbitrary code execution? De-anonymization? They then go on to say that they haven't told the Tails maintainers what the vulnerabilities are, but will "in due time", implying they're going to sell them off to the government first. Exodus Intelligence also does a lot of business with the US government, possibly including the NSA.
To me, this sounds like they probably found some minor zero-days and are trying to spread FUD (likely spurred on by their clients in the government) to get people to stop using Tails. After all, we know that the NSA is trying to put people who attempt to download Tails on a watchlist for further scrutiny.
The kind of environment where the attacker is a sysadmin with access to the box and the ability to do whatever they feel like with BIOS, including enabling USB boot.
The default security posture of most organizations these days is to assume that a trusted insider will exploit the system at some point. Therefore everyone is implementing damage mitigation techniques so that they can respond quickly and understand the scope of the inevitable breach when it does occur.
Everyone is watching everyone else. The security guys get access to the firewalls and the IDS, but cannot touch the servers. The server guys cannot touch the backups. The backup team cannot initiate a restore without two levels of change control approval. It is a serious PITA for everyone involved and a gross inefficiency.
The first time an auditor told me that they cannot trust me, my knee jerk reaction was to tell them to go fuck themselves. Eventually I realized that I am in a very risky position with access to a lot of sensitive information. The key is not that they do not trust me, it is that they CANNOT trust me. While I may be trustworthy, who is to say that someone else in my same position, with my same level of access, is also trustworthy? Just like I have to assume that any executable downloaded from the internet is potentially full of malicious code, the risk management folks have to assume that every sysadmin in the organization is potentially full of malicious intent.