NOTA BENE

Notes, comment and buzz from Eugene Kaspersky – Official Blog

February 26, 2015

Cancunference 2015.

Some ten-plus years ago, our then still quite small company decided to push the boundaries – literally: we went transnational. Before long we found we had expert-analyst KLers working in all corners of the globe, all of them communicating with one another by email, messengers, telephone and other indirect means. Nothing wrong with that really, but still, it’ll never beat face-to-face interaction. So we decided to have a yearly jamboree where we’d all get together and top up on the much needed proper face time. That was when our annual conference for IT security experts was born: the Security Analyst Summit (SAS).

cancun-mexico-sas2015-1

cancun-mexico-sas2015-2

Read on: Work hard, play hard, like always…

January 16, 2015

Encrypted communications and real-world security: finding a balance

The latest debate that followed David Cameron’s proposal to ban encrypted personal communications in the UK has raised several very important issues.

The proposal would include a ban on messaging services like WhatsApp, iMessage or Snapchat in the UK. Technically this is possible to do, however such a ban on using all encrypted communication channels is not easy to enforce.

And I doubt that it will actually bring significantly more security to offline UK.

The mandate of the security services and law enforcement agencies is to keep the general public safe from criminals, terrorists and all sort of other threats. It seems that the security services want to be able to access our communications in order to be able to stop and prevent illegal activities and, ultimately, better protect people.

Encryption is vital for cybersecurity; it’s used first and foremost to keep communications safe from hackers and cybercriminals.
Do we need to give up the protection of our our data and online communications in order to improve real-world security? I seriously doubt we should.

I think that, if implemented, a ban on the use of encryption in online communication will not tangibly increase offline security. But it will definitely damage the state of cybersecurity and ultimately expose ordinary users as well as businesses to all sorts of cyberattacks, hacks and espionage.

Governments have made attempts to compromise cybersecurity to gain intelligence. For example, we have already seen government-grade malware, such as Flame, exploiting legitimate software, such as Microsoft Update, among other things.

I don’t know the value of the intelligence they obtained during this operation, but the existence of such malware did not contribute positively to global cybersecurity.

I think the real problem here is that global leaders and security services apparently see a contradiction between security and cybersecurity; while the latter should in fact be an integral and valuable part of the former.

October 21, 2014

Geography lesson.

Every day we release up to 2000 updates for our products.

Every week our users around the globe download those updates over a billion times.

Every month we distribute around four petabytes of updates.

These updates (together with our other technologies) protect you against new cyberthreats. In recent years we’ve been seeing new malware popping up not just every day or every hour, but every minute and even every second! Each year we analyze more than a billion samples of malicious code.

For the average user, receiving antivirus updates is a simple, automatic process. They run silently in the background without disturbing you (and quite right too). However, there’s a lot more to an update than first meets the eye. Updates are merely the tip of a sophisticated iceberg that connects our products to a huge distributed IT system that we built up ourselves using a whole bunch of original ideas and know-how.

That’s the overall scheme. The details get more interesting…

Kaspersky Internet Security Update

Read on: So what actually happens when you update your antivirus?…

September 29, 2014

The evolution of OS X malware.

Is there any (Mac) OS X-specific malware around?

Oh yes. But for some odd reason I haven’t said anything interesting on this topic for quite a while…

The last time was two and a half years ago. Yes, that’s how long it’s been since the global Flashback worm outbreak that infected 700 thousand Macs worldwide. The security industry made quite a bit of noise about it (and quickly disabled the Flashback botnet), but since then – mostly silence… It might seem to some that ever since there’s been a complete lull on the Mac-malware front and not one bit of iMalware has disturbed Apple Bay’s calm waters…

But they’d be wrong…

Mac malware is not amyth, they do exist

Sure, if you compare the threat levels of picking up some malware on different platforms, at the top of the table, by a long way, as ever, is the most widely used platform – Microsoft Windows. Quite a way behind it is Android – a relatively new kid on the block. Yep, over the past three years the cyber-vermin has been seriously bombarding the poor little green robot with exponentially increasing levels of malicious activity. Meanwhile, in the world of iPhones and iPads, except for very rare cyber-espionage attacks, there have been hardly any successful attacks thereon (despite using various exotic methods). It’s a similar story with Macs too – things are relatively peaceful compared to other platforms; but of late there have been… stirrings – about which I’ll be talking in this post.

Briefly, a few numbers – kinda like an executive summary:

  • The numbers of new for-Mac malware instances detected in the last few years are already in the thousands;
  • In the first eight months of 2014, 25 different ‘families’ of Mac malware were detected;
  • The likelihood of an unprotected Mac becoming infected by some Mac-specific-unpleasantness has increased to about three percent.
In 2013 alone @kaspersky detected ~1700 malware samples for OS XTweet

Read on: let’s dig deeper and look at the situation from a malware expert PoV…

September 2, 2014

Under the hood – 2015.

We’ve a tradition here at KL (besides the summer birthday bashesNew Year shindigs and the rest, that is). Every summer we launch new versions of our home products. Er, and it’s already the end of summer! (Eh? Where did that go?) So let me give you the highlights of the juiciest new features of our 2015 versions, or, to put it another way – about the latest sly tricks of the cyber-villains that we’ve successfully been busting with our new tech that’s winding its way into KL-2015s :).

All righty, off we go…

Kaspersky Internet Security 2015 - Main Window

What’s new in Kaspersky Internet Security 2015? @e_kaspersky reportsTweet

Read on: The all-seeing eye of Sauron. No more…

July 26, 2014

Cybernews from the dark side – July 26, 2014.

Remote controlled car – your car, while you’re driving it…

News about new hacks, targeted attacks and malware outbreaks is beginning to bore the general public. It’s becoming an incessant stream after all. What isn’t boring the life out of the general public is something a bit more unusual: stuff you wouldn’t dream could be hacked… getting hacked.

A report from China told how hackers broke into the Tesla motor car’s gadgetry – as part of a contest during a hacker conference. So, why Tesla? What’s so good about Tesla? Well, that’ll be its being an electric car, and its being crammed with so much ‘smart’ electronics that it hardly resembles an automobile than a mobile supercomputer. Still, what was Tesla expecting? Any new functionality – especially that developed without the involvement of IT security experts – will inevitably bring with it new threats via vulnerabilities, which is just what the hackers at the conference in China found.

Cybernews from the darkside

Read on: malware getting closer to industrial systems…

July 14, 2014

Our antivirus formula.

Every system is based on a unique algorithm; without the algorithm there’s no system. It doesn’t really matter what kind of algorithm the system follows – linear, hierarchical, determined, stochastic or whatever. What’s important is that to reach the best result the system needs to follow certain rules.

We’re often asked about our products‘ algorithms – especially how they help us detect future threats better than the competition.

Well, for obvious reasons I can’t divulge the details of our magic formulae; however, what I will be doing in this tech-post (perhaps the techiest post on this blog ever) is open ajar the door to our technological kitchen – to give you a glimpse of what goes on inside. And if you still want more info, please fire away with your questions in the comments, below.

Read on: A very brief look at our Coca-Cola-like ‘secret’ magical formula in a little over 2000 words…

July 3, 2014

Beyond good and evil?

A few days ago Microsoft announced a large scale raid on the dynamic DNS service No-IP, as a result of which 22 of its domains were seized. The guys in Redmond said there were very good reasons for this: No-IP hosts all kinds of unpleasant malware; No-IP is a breeding ground of cybercriminals; No-IP is an epicenter for targeted attacks; and No-IP never agrees to working with anyone else on trying to root out all the badness.

Like in most conflicts, the sides have exchanged the contradictory volleys of announcements in the eternal tradition of ‘it’s his fault – no she started it’.

In particular, No-IP has said it’s a real goody-two-shoes and always willing to cooperate in eliminating sources of cyberattacks, while its clients are most displeased with the raid and consider it an illegal attack on legal business – since it’s possible to find malware practically anywhere, so interrupting services through a court is simply not on.

Is it legal to shut down a service because of #malware found?… When it can be found everywhere?…Tweet

In the meantime, the result of the raid has been rather far-reaching: more than four million sites were pulled, including both malicious and harmless ones – affecting 1.8 million users. Microsoft is trying to sieve the wheat from the chaff and get the clean sites back up and running; however, many users are still complaining about ongoing disruption.

To work out who’s to blame is a thankless and probably hopeless task. I’ll leave the journalistic investigations to… the journalists. Instead, here let me give you some food for thought: dry, raw facts and figures – so maybe/hopefully you’ll be able to come to your own conclusions about the legality and ethicality of MS’s actions, based on those facts and figures…

1)      Shutting down 22 No-IP domains affected the operations of around 25% of the targeted attacks that we keep track of here at KL. That’s thousands of spy and cybercriminal operations ongoing for the last three years. Approximately a quarter of those have at least one command and control center (C&C) with this host. For example, hacker groups like the Syrian Electronic Army and Gaza Team use only No-IP, while Turla uses it for 90% of its hosts.

2)      We can confirm that out of all large providers the No-IP dynamic DNS was the most unwilling to cooperate. For example, they ignored all our emails about a botnet sinkhole.

3)      Our analysis of current malware shows that No-IP is often used by the cyberswine for botnet control centers. A simple search via the Virustotal scanning engine confirms this fact with a cold hard figure: a total of 4.5 million unique malware samples sprout from No-IP.

4)      However, the latest numbers from our security cloud (KSN) show something not quite so cut and dry. Here’s a table showing detections of cyberattacks from dozens of the largest dynamic DNS services:

Service % of malicious hosts Number of detections (in a week)
000webhost.com 89.47% 18,163
changeip.com 39.47% 89,742
dnsdynamic.org 37.04% 756
sitelutions.com 36.84% 199
no-ip.com 27.50% 29,382
dtdns.com 17.65% 14
dyn.com 11.51% 2321
smartdots.com 0.00% 0
oray.com 0.00% 0
dnserver.com 0.00% 0

So – No-IP isn’t leading in the number of detections, even though they’re still really high compared to most.

Here’s some more info for comparison: the % of malware hosts in the .com zone makes up 0.03% of the total; in the .ru zone – 0.39%; but in No-IP the figure’s 27.5%!

And now for other figures that add a bit of a different perspective: in one week, malware domains on No-IP generated around 30,000 detections, while in the same week on one of the most malicious domains in the .com zone, the figure was 429,000 – almost 14 times higher. Also: the tenth most infected domain in the .ru zone generated 146,000 detections – that is, about the same as the first ten providers of dynamic DNS mentioned above put together!

To summarize…

On the one hand, blocking popular services that are used by thousands – if not millions – of typical users: it ain’t right. On the other hand, closing spawning grounds for malware is right – and noble.

The takedown of No-IP domains. Was it right or wrong? Ambiguity with a big ATweet

But then mathematics takes on the role of devil’s advocate, and proves:

Quantitatively, closing all the domains of No-IP is no more effective in combatting the distribution of malware than closing one single top malware domain in one of the popular zones, i.e., .com, .net, or even .ru. Simpler put, even if you were to shut down all providers of dynamic DNS – the Internet still wouldn’t become ‘cleaner’ enough to notice the difference.

So there you have it – ambiguity with a big A. 

It leaves anyone in their right and honest-with-themselves mind to admit things are far from black and white here, and as regards the right and wrong, or good and bad, or Nietzsche’s thing – who can tell?

Still, another thought comes to mind at some point while reflecting on all this…

It’s further evidence that as soon as the quantity of piracy or degree of criminality gets above a certain threshold, the ‘powers that be’ get involved all of a sudden and start closing services, ignoring any notions of Internet freedom or freedom to do business. It’s just the way things are, a rule of life of human society: If it stinks, sooner or later it’ll get cleaned up.

The list of blocked services is already rather long: Napster, KaZaA, eMule, Pirate Bay and so on. Now No-IP‘s been added to the list.

Who’s next?

// Bitcoin? It’s already begun.

 

June 30, 2014

Cybernews from the dark side: June 30, 2014

Stock market hacks for microsecond delays.

Cyber-swindling gets everywhere. Even the stock market. First, a bit of history…

The profession of stockbroker was once not only respected and honorable, but also extremely tough. Dealers in stocks and shares once toiled away on the packed floors of stock exchanges and worked silly hours a week, stressed to the limit by relentless high pressure decisions all day (and night). They bought and sold securities, stocks, bonds, derivatives, or whatever they’re called, always needing to do so at just the right moment while riding the waves of exchange rates and prices, all the while edging nearer and nearer to serious heart conditions or some other burn-out caused illness. Other times they simply jumped out of windows to bring a swift end to it all. In short – hardly the world’s best job.

Anyway, all that was long ago. All that hard manual labor has been replaced by automation. Now thinking hard, stressing and sweating aren’t needed: a large proportion of the work today is carried out by robots – special programs that automatically determine the very best moments to buy or sell. In other words, the profession of stockbroker has in large part been boiled down to the training of bots. And to these bots reaction times – to the microsecond – are vital to take advantage of this or that market swing. So speed literally depends on the quality of an Internet connection to the electronic stock exchange. That is, the nearer a robot is physically located to the exchange, the higher its chances of being the first with a bid. And vice versa – robots on the periphery will always be outsiders, just as will those not using the very latest progressive algorithms.

These critical reaction times were recently tampered with by unknown cyber-assailants. A hedge fund’s system was infected with malware to delay trading ability by a few hundred microseconds – which can – and probably did – make all the difference between clinching deals and losing them.

bae-600x255

Read on: Your password for a Twix?…

June 15, 2014

10 years since the first smartphone malware – to the day.

On June 15, 2004, at precisely 19:17 Moscow time something happened that started a new era in computer security. We discovered the first malware created for smartphones.

It was Cabir, which was infecting Symbian-powered Nokia devices by spreading via unsecured Bluetooth connections. With its discovery the world learned that there was now malware not just for computers – which everyone already knew too well about (save for the odd hermit or monk) – but also for smartphones. Yes, many were scratching their heads at first – “viruses infecting my phone? Yeah, pull the other leg” – but the simple truth of the matter did finally sink in sooner (= months) or later (= years a decade!) for most people (some still aren’t aware). Meantime, our analysts made it into the history books!

Why did we christen this malware Cabir? Why was a special screened secure room created at our Moscow HQ? And how did Cabir end up in the pocket of an F-Secure employee? These and other questions were recently put to Aleks Gostev, our chief security expert, in a interview for our Intranet, which I thought I’d share with you here; might as well have it from the horse’s woodpecker’s mouth…

Incidentally, the story started really running when we used these two devices to analyze the malware:

The legendary Symbian-powered Nokia phones we used to analyze Cabir

…but more about those below…

Read on: An unusual file n the inbox…