« first day (3043 days earlier)      last day (1858 days later) » 

2:22 AM
Anyone here fuzzed ClamAV before?
I'm digging up my notes from old projects and see a long-abandoned project I had to add a sandbox to ClamAV to mitigate its nastier security vulnerabilities.
I think I dropped it because ClamAV is a crappy AV anyway.
 
2:42 AM
@forest, agree. Tried Clam before and dropped it. Not to mention the vulnerabilities but the performance of it...
Actually my workplace is moving to scan less next gen AV that uses machines learning technology, and less signature detection methods
 
I.e. buzzword snakeoil. :P
(And heuristic detection has been the norm for over a decade anyway)
I've only used ClamAV on a live Linux DVD to scan Windows systems.
So I never really needed it to be fast.
 
So you are in favor of heuristics signature detection?
 
Well when AV is absolutely necessary, yeah, but I don't really like AV.
Heuristic detection is often superior to signature detection.
 
2:59 AM
Agree - polymorphism of malware is a big point in favor of behavior based models of AV
A virus changing its footprint (signature) so rules no longer recognize it...
 
Yep. Of course, even heuristics analysis is trivial to defeat.
In the end, heuristics analysis is only good at catching badly written malware.
And signature analysis is only good at catching known malware.
 
So that leaves the well - crafted 0 days
 
Nah, just decently-crafted malware. No need to use a 0day exploit.
 
Another point is that the best of technology may be not sufficient for malware that relies on social engineering, deceiving the user psychology to execute it...
 
Yep, that's another big one, especially when it comes to phishing (no need to execute malicious code outside the browser if all you need is creds and a dumb user).
 
3:03 AM
A Trojan horse may be a good example here
 
I mean I'm not even an expert malware developer, but I could trivially write useful malware that cannot be detected by AV (at least until it is discovered).
 
on the point of phishing, have been getting some interesting answers on my recent question form yesterday. Most seem to object to spear fishing as been excessive
 
I honestly think even those kinds of extreme tests are fine.
The biggest issue is when there's no way to distinguish a genuine email from a phishing one, suggesting genuine emails need to be changed.
 
thought so from your comment "Know the enemy..."
Mind sharing a little more details, without degrading reality of these tests?
 
Primarily by ensuring that legitimate emails never ask for sensitive information.
There are other ways to obtain sensitive information legitimately but securely.
It's a big problem when real emails require you blindly trust them.
Compare that with companies which tell you repeatedly that they will never ask you for password information, etc. and that anyone asking you for it is lying to you.
There's a term, scamicry, which is when real message seem to imitate scams.
Can't remember where I first saw that term, but it's also in this pdf.
 
3:21 AM
Agree...with blindly being the key word here. Show us that you are legit, don't just the end to trust that you are :)
user*
 
Mutual authentication is always vital.
 
We are lucky that they are methods such as digital signatures and DMARC, SPF etc. that allow authentication happen
And on topic of spear phishing, one successful phish at my job was actually whaling
Mind sharing what your company is doing for phishing or AV?
 
 
6 hours later…
9:39 AM
@forest why is that? I have no experience with it, but a colleague told me about it and it sounded kinda nice
 
 
1 hour later…
11:02 AM
@forest some organisations I worked with did great work on phishing scenarios, reducing the click thru, improving education for those who clicked one 2 or more tests...and then they sent marketing emails from external bodies, with links, and got annoyed when people didn't respond...
 
 
1 hour later…
Anonymous
12:26 PM
@TomK. Yeah for the £10 a month Tom, HTB VIP is for sure worth it if you're going to be using the platform a lot.
 
Anonymous
12:49 PM
@forest Some interesting news for you; I did an article for SecJuice about the UK porn ban and AgeID (who are one of the many companies owned by MindGeek, who are also supplying an age verification tool to the porn industry for this ban) emailed the editor at SecJuice and asked us to remove a section from our article and replace it with a modified version which they've written themselves haaha.
 
6:08 PM
@J.J wow
what was your response
 
 
1 hour later…
Anonymous
7:23 PM
@RoryAlsop We virtually told them to get fucked.
 
9:14 PM
Quite right!
 
Anonymous
@RoryAlsop I'm waiting for them to start threatening legal action, I'm sure that is where this is going.
 
Do you think you wrote anything that could be legally challenged? Or are they just being stroppy
 
Anonymous
9:34 PM
@RoryAlsop Two seconds, you tell me. This is the statement they're unhappy with.
 
Anonymous
I feel a lot happier about this method of verification as opposed to sending a photo of my ID and tying my porn verification account to my real identity. Despite MindGeek claiming that AgeID won't track you & what you watch, given the track record of companies saying such statements, it makes it hard/concerning for us to believe that they won't somehow use this data in ways which they should not.
 
Anonymous
Now, they're saying this section Despite MindGeek claiming that AgeID won't track you & what you watch, given the track record of companies saying such statements, it makes it hard/concerning for us to believe that they won't somehow use this data in ways which they should not. should be changed, apparently we're incorrect which we may be.
 
Anonymous
However, no where do we claim that is fact, you can clearly see that statement is an opinion we are fully entitled to doubt them and there is no crime against doubting something and writing about it. I'm no expert but slander needs damages, this doesn't cause any.
 

« first day (3043 days earlier)      last day (1858 days later) »