Monday, September 12, 2016

Cyber, just as old as ancient human history

Recently I visited a seminar on which the question was asked about what was new on the phenomenon cyber. Although I somehow find we still should #ditchcyber, I started thinking about that question. After some internal computing time I came to the conclusion that cyber is nothing new and that is just as old as ancient human history. Well, that is, the paradigm in which cyber resides is ancient.

Fifth paradigm of warfare

In June 2016 the NATO officially declared the cyberspace as the fifth domain of warfare. The other four are land, sea, air and space. With the ongoing attacks and governments ranking up their cyber-capabilities, our precious infrastructure is becoming ever more a potential zone of conflict. But why is this a development that needs real attention?

Every paradigm strengthens the other

Once there was merely conflict on land. People attacked each other with ordinary weapons. First it was all melee combat which ‘soon’ was strengthen by ranged combat. But still, everything was done on land and defenses grew in time to withstand such attacks. When the second paradigm, sea, came into play battles changed quickly. Not only were battles fought on sea, but sea was also used to strengthen land combat. Troops could be sent in through ships in (at first) defenseless harbors.

The third paradigm took a while to arise in our arsenal of conflict zones, but it came with devastating capabilities. Through the air many defenses became almost pointless (like city walls) and aerial combat strengthen both land and sea warfare. Even planes could take off from carriers to strike on land, sea and in the air.

Space was the fourth paradigm, and as far as we know all countries uphold the international treaty about not bringing warfare to space. Often people do not know that the treaty is only about ground-to-space and space-to-space combat and that it does not include space-to-ground combat. So it is ‘allowed’ to use satellites for ground bombardments, but you will have to violate the treaty to take down the satellite that is attacking.

We should be thankful for countries upholding the treaties so far. When you look at the four paradigms combined with our increasing capabilities you see that the potential casualties of conflict increase. And I want to emphasize on the potential part, because since 1945 the absolute casualties due to conflict has been decreasing ever since (Our World in Data).

Screenshot of Kaspersky Lab Cybermap
Cyber, the fifth paradigm, recently emerged from our endless increasing capabilities in computing, storage and networking power. The Internet came to life and our lives and everything else are becoming ever more interconnected. But also already present land, sea, air and space capabilities are strengthened by the use of cyber technologies. Just think about drones that are bombing regions remotely and critical infrastructures like power grids that are taken offline through the means of cyber. The potential damages (most often economical for now) are increasing all the time. And sooner, rather than later, casualties will also be the result of cyber-attacks.

Casualties just might seem far-fetched, but think about remotely interrupting pacemakers and taking down critical infrastructure like electricity and fresh water. Taking does down in regions suffering from extreme heat might result in many fatalities due to dehydration and overheating. But also driverless cars crashing into each other, or planes that can be hacked from the ground.

Cyber is not new, but it has a key new characteristic

Most often our government is taking care of the defense of land, sea, air and space. There is an army for external intruders, and there is often a police-force for internal advisories. Countries also have intelligence agencies that feed the governments they represent with intel on what (potential) enemies our doing and planning. We as citizens might have a good night sleep without worrying to much about being invaded, knowingly that this sadly does not apply to all of us.

The new aspect of the cyber domain is that we cannot depend on the government alone for proper protection of, well, kind of everything and more than that. Companies and individuals alike also have to contribute to the overall safety of our world. It is imperative that those with power use such powers with responsibility. And I do not mean superheroes, but nation leaders, CEOs, CIOs, CTOs, CISOs and everyone else that can influence budgets to rank up the cyber defense. And in essence, they just might be the modern-day cyber-heroes.

So is Cyber new? No, it’s paradigm is ancient, just like any other. It has opportunities, risks, weapons and defenses. But the fact that it needs to be protected by everyone, instead of only the government, with the power to influence its security is new.

Oh, and again #ditchcyber, it most often clouds (no pun intended) discussions on the things that matter.

Wednesday, August 17, 2016

Make Security and Privacy Awareness ubiquitous

"Yeah, let's create a page and put all information there that our users and/or customers need to stay secure!” Sounds familiar? Do you have a so called awareness page somewhere on your intranet or website? But are you suffering from lack of traffic, or at least, a lack of success of that page?


In my recent post “Need Security Awareness? You're doomed!” I talked about that Security Awareness is the last thing you should focus at (I was overstating that on purpose of course). For the worse part of it, with awareness you are depending on the weakest link in the chain, and that are humans. And humans have proven to be relentless in not following guidelines whenever they feel they need to. So how we can inspire to follow the guidelines? Well, with proper awareness...

I think that there are two fundamental principles that needs to be taken into account with awareness. First of all, do not only tell how, but focus on why. And when you tell it, tell it when it happens. Tell it when there is a change! I will zoom in on those two principles, but first I want to make another difference. There are also two types of awareness. One is about Security and the other is about Privacy. A good understanding of the differences of these two types will help you in telling why and how.

Example with thanks to Bob

A customer, let’s name him Bob, logs on your website and decides he wants to change his password. He goes to his accounts settings and to the tab about his passwords. He just sees two blank fields, fills in the same password as his e-mail account twice and hits the button OK. Some algorithm approved his password, as it is complex enough. Say something like: Th!s!sN0tMyP@ssw0rd. It is long and difficult. Right?

Two things failed here. First of all, he re-used his password and his password is anything but complex. It is probably far from ‘unique’ (for as far that it is literally possible) and likely ill-often used and present in password-guessing tools. But there is a third thing that failed tremendously. The chance for an effective awareness message presenting to Bob has not been given.

What if there was a message like below, before Bob could change is password?

When choosing your password it is important that you choose one that you do not use with other services. It is possible that your password gets stolen through another service than ours, but that it is being misused on our service. Your privacy is then compromised and that is something we really want to prevent. To remember all your passwords you can use a password manager tool which will help you better protect your own privacy. Click here to see our central privacy protection page to learn more.

And then present the form to change the password. Chances are that he will not re-use a password. It is not a 100% guarantee, but it has higher potential than saying nothing. Why? Because you tell why (it’s about privacy), you tell how (that’s about security) and you tell it when it matters (when the change happens)! I strongly state that this type of awareness is more effective than a distant page.

But what about Th!s!sN0tMyP@ssw0rd ?

And this is where we need technology to help our colleagues and customers (instead of failing awareness). Implement good filters and regular expressions to enforce a good password policy, but also check it against often used passwords. Also prevent (whenever possible) the use of compromised passwords in combination with the name of the account. Just do not tell the user how a password should look like, but help him or her with it. It’s far more effective!

I am not going into the debate now what the complexity of a password should be. But I rather have it unique and long, than complex and shorter. From a computing perspective, no combination of characters is more complex than another combination of characters. The password ‘A Purple Bunny is swimming in the Ocean’ is more likely to be secure than the example of Bob. Why? Because it makes no sense to people building brute-force algorithms and it has more characters. And it is easier to remember and therefore chances of it being written down are slimmer.

Next step is deleting the awareness page?

Now I am not stating we all should deleting our awareness page, and move to a system like a described above. It’s smart to have a page which contains all the important details combined together. Sometimes people do get interested and it would be a wasted opportunity to not satisfy their information hunger. Let these two co-exist, but focus your time and energy in making awareness training ubiquitous. Make it present everywhere it matters.

Oh, and try to avoid the word awareness. It says something about a person not having something (you are not aware!) then that it is about gaining something (better protection of their privacy).

In other words

Make awareness training ubiquitous by incorporating it within your entire environment by telling how, why and by telling it precisely on the moment it has the greatest impact. The place where change is done, the place where it matters.

Sunday, July 17, 2016

Single Sign On, potentially your biggest Security headache

It is not uncommon for companies pursuing the principle of Single Sign On (SSO) for their information systems, often disguised with the claim to ‘improve’ security. Although I agree that the usability for users are increased in almost in every case, I do not agree that the same applies towards security. On the contrary, if not implemented in a good manner it decreases your security and increases your headache!

To explain my statement, I will take you through some layered thinking about the subject and address some things to do to address SSO in a secure manner.

Password ethics

The ethics towards password use are most often not that of the standards that we as Security Officers would like to see it. This concerns many aspects, such as password uniqueness, password length, and password secrecy. Passwords (or passphrases for that matter) are often not unique, are to short and are kept secret in a rather insecure manner.


Imagine an IT-environment with full-blown SSO with primary accounts that have the same password as somebody’s home computer, combined with passwords that are way to short (anything less than 12 positions is short) and are kept in an spreadsheet on a computer without disk-encryption. And now imagine that such password gets compromised and all extremely sensitive data can be accessed on your corporate network through that one account, just because you implemented SSO.

When thinking about SSO, you need to re-think your password ethics.

Password uniqueness

It is hard to address uniqueness and it is for the most part pure awareness of the people. But there are some things you can address. First of all, make sure passwords cannot be re-used. Whenever a user changed his or her password, it should be at least 20% new. When considering a password of 12 characters, you already have to change at least 3 positions.

Also audit the passwords of your users on regular intervals. On the Internet you can find sources with the most often used passwords. Take a big amount (for instance, the 20, 30 or 100) of most used passwords and make sure users cannot use those passwords and if they do, forcible reset their accounts. Apply these tactics also when password databases of other companies get compromised and compare (if possible) those passwords against your database to prevent misuse of accounts due to hacks at other companies.

Password length

Password length is rather easy to fix. First of all, bump it up from the pretty much default of 8 positions to 12 positions. There is some debate about whether or not it should be a difficult password with special characters and all. I tend to say that when your password is unique within the scope of all accounts of yourself and that it is at least 12 positions long, you are good (enough).

Password secrecy

And now password secrecy. Who doesn’t know that one guy who saves all his passwords in an spreadsheet, so it is convenient for accessing them? Do you realize that you’ll probable have some of those as your co-workers? And that they are likely have access rights to important data-sources?

So give your users the tools to safely store every password. Whether or not this is an on-premise tool or a Cloud-based tool, make sure it has enterprise features like central management for your admins and a full audit trail for your auditors. It will make everybody’s life easier and users can keep their passwords safe. The fact that you are on the verge of implementing SSO, does not mean you can skip the password manager.

Data classification ethics

And then there is the aspect of data classification. The ethics of topics such as these can be made extremely complex or extremely simple (to simple). In many organizations data classification is not really done and for the better part of it, I can totally understand that. But for an improved SSO you will need to have some form of classification.

If you have data classification, you can skip this part. Otherwise this might come in handy. When there is a lack of data classification you can approach this subject with three types of classifications. The first is public.

Public data

Public data may be read by at least anyone in the organization. It does not mean it needs to be readable to everyone, but it might be just as well.

Corporate data

The second one is corporate data. Corporate data is data that may not be read by everyone in the organization. This type of data is important for the business to function, but it lacks protection of laws and it is also not intellectual property.

Secret data

The third one is secret data. All personal identifiable information (PII), intellectual property (IP), and all other information that is subject to law or regulators must be considered as secret data. Most often of the time you can find the crown-jewel data in this category. Think about information that is vital to stock-trading and may not leak due to regulators and think about personal information of your customers that is subject to the EU Privacy Directive.

And now grab some engineers in a team and divide the information systems into these categories. I bet that in 90% of the time these assessments can be done out the top of the head of the engineers.

Stepping up your SSO

When you have put the controls in place to improve password uniqueness, length and secrecy you can move forward to implementing SSO. When implementing SSO you really need to factor in the two-factor authentication (2FA). And this principle is easy.

Whenever a user accesses information in a higher level category, you will need to ask for a new 2FA-code.

There is no need for re-asking the password, just the code from the authenticator app, physical token, YubiKey, SMS or whatever you use for 2FA. This mechanism prevents that when a password gets compromised, that all data can be accessed without further obstruction. 2FA also helps users to detect (in case of SMS) login attempts on their accounts.

There is on caveat though…

There is one caveat in SSO that is often overlooked. The moment you start using a system (such as Active Directory) as the primary source for identity management, it is then by definition the most critical system in your network. You really need to be rigorously protective towards that system.

Therefore, never allow unencrypted traffic towards and from such central system. Never use insecure APIs to connect and make obsolete technologies impossible to use. All security controls are completely irrelevant when some old system that is not SSO compatible transmits the username and password unencrypted over the network just to get it to work.

Summary

To implement a secure single sign on (SSO) you will need to improve the password uniqueness, length and secrecy. You will also need to have at least three categories of data classification. With these controls you can implement SSO with the use of two-factor authentication (2FA). Whenever a system in a higher level of category is accessed, the principle of 2FA needs to be applied. And harden the central system with all its identities and make sure that all data exchanges are well secured.

If you have done the above, you are a big step further to lessen your headache.

Sunday, June 26, 2016

Boek review: Komt een vrouw bij de h@cker, van Maria Genova (Dutch)


Een poosje geleden zag ik wat tweets over het boek "Komt een vrouw bij de h@cker" van een ene @genova2 voorbijkomen. Hoewel ik dacht dat het boek in de categorie fictie viel besloot ik Maria Genova (de schrijfster van het boek en de persoon achter de hiervoor genoemde Twitter-handle) te gaan volgen. Toen ik enkele maanden later op de conferentie van (ISC)² SecureNetherlands 2016 aanwezig was zag ik in de agenda dat Maria kwam spreken over haar boek. Na een boeiende talk kwam ik met haar via Twitter in contact en kocht ik via haar een persoonlijk signeerde boek! Oh trouwens, het boek is helaas allesbehalve fictie...

Komt een vrouw bij de h@cker gaat over identiteitsfraude. Ze beschrijft waargebeurde cases ten aanzien van het misbruiken van gegevens door anderen om zich voor te doen als jou. Meestal met nadelige financiële gevolgen voor de originele identiteit (lees: jij). Situaties zoals fraude met betrekking tot het inschrijvingen op je woonadres, mobiele telefoonabonnementen, bestellingen bij online winkels, misbruik van bankgegevens en ga zo maar door, komen allemaal aan bod. Ook gaat ze in op overheidsinstanties en hoe slordig daar (soms?) wordt omgegaan met gevoelige persoonsgegevens. Vaak overigens op de meest eenvoudige manier, zonder complexe hacks door hackers.

Dit is allemaal bijna niet voor te stellen en dus besloot Maria om contact te leggen met een echte hacker om zodoende uit eerste hand te zien hoe eenvoudig digitaal inbreken is. En de informatie op onze computers laten weinig tot de verbeelding over, want nagenoeg alles is te vinden op onze computers. Denk hierbij aan volledige kopieën van identiteitsbewijzen zoals paspoorten en rijbewijzen, maar ook accounts en wachtwoorden die we opslaan in eenvoudig te lezen bestanden, en intieme en soms ook pikante foto's. En niet te vergeten de slechte discipline die gehanteerd wordt ten aanzien van het installeren van alle updates en het hebben van een malware (virus) scanner. Het is een snoepwinkel voor de digitale inbreker.

Dat identiteitsfraude bestaat en plaatsvindt kan ik me nog wel mee verzoenen. Criminaliteit bestaat nu eenmaal en hoewel we het zeker moeten bestrijden, moeten we niet de illusie hebben dat we het kunnen uitbannen. Niet op korte termijn in ieder geval. Wat ik zelf gewoon bijzonder vind en eigenlijk gewoon weg niet begrijp is de mate waarin een slachtoffer wordt ondersteund in dit proces (althans, de mate waarin hij of zij niet wordt ondersteund).

Op het moment dat we iemand op straat zien overlijden krijgen we slachtofferhulp aangeboden (sterker nog, ik kreeg het zelfs aangeboden toen ik aangifte kwam doen van diefstal van mijn fiets). Maar slachtofferhulp bij identiteitsfraude gebeurd gewoon niet. Vaak krijg je al niet eens een aangifte goed verwerkt, is de politie niet deskundig (genoeg) om je te ondersteunen en heb je vaak de schijn al tegen want het is toch 'jouw' handtekening!

De financiële schade is vaak al niet te overzien. Van enkele duizenden tot vele tienduizenden euro's die je gewoon kan ophoesten omdat je in de rechtbank niet geloofd wordt. Onschuldig gevangen zitten is helaas hierbij ook geen uitzondering! Maar ook de impact op je persoonlijke leven is voorbij het denkbare. Geliefden die je niet meer geloven en van je gaan scheiden, vrienden en familie die je in de steek laten want "waar rook is, is vuur" en eveneens het verliezen van je baan en huis.

Identiteitsfraude is realiteit, het krijgt te weinig aandacht van de overheid en justitie, de pakkans is klein en de hoeveelheid aan onschuldige slachtoffers groot! Het boek samengevat in één woord: doodeng...

Maria gaat gelukkig ook in op wat eenvoudige en toepasbare oplossingen die al een groot verschil kunnen maken. Denk hierbij aan het hebben van een uniek wachtwoord voor elk online account, geen kopieën van je legitimatie meer verstrekken tenzij dit van de wet moet (meer info hier), terughoudend zijn met welke informatie je deelt en het installeren van alle updates voor je computer, tablet en telefoon. De tips beslaan meerdere pagina’s en ze zijn erg nuttig en helpen allemaal een beetje bij het (hopelijk) voorkomen van identiteitsfraude.

Wat mij betreft een must-read voor iedereen, ongeacht leeftijd, opleidingsniveau, werkniveau en geslacht. Want iedereen kan slachtoffer worden!

Eerste publicatie: 2014
Pagina’s: 224
ISBN: 978-9-089-75292-5
Link: mariagenova.nl
Lezing: mariagenova.nl/lezingen

Friday, June 17, 2016

My talk at US Consulate about Cyber Security and Agile Development

I had a talk at the United States (US) Consulate in Amsterdam last Tuesday (June 14, 2016) on a Cyber Security v. Agile eCommerce event. Companies like SBS Broadcasting, KLM, ION-IP, WhiteHat Security, Bureau Brandeis, Isatis Group and many more attended this event. I was asked by ION-IP to speak at this event and, of course, I immediately said yes!

Before I say something about my talk, let me first start with the US Consulate itself. This was a very nice and new experience for me, especially from a Security perspective because it has airport-tight Security levels. First of all, I needed to get a personal invite the Commercial Specialist of the US Commercial Service of the consulate itself (besides the invite by ION-IP and WhiteHat Security). No invite from the consulate itself means no access. For obvious reasons I had to show my passport (driver’s license was also possible) and had to turn over all my electronic devices.
US Consulate in Amsterdam - Source: Wikipedia
Thinking to be efficient, I had already switched of my phone, but I had to turn it on again. The reason for this was so they could see it was actually a phone. Then it was tested for drugs and explosive substances, and then I had to switch it off again and turn it over. I decided to not bring my smartwatch, because that would have been submitted also. The next step was to walk through a detector and, thankfully, I had to only take of my belt and no other clothing. We were then escorted by an employee towards to meeting room (through a couple of locked doors). We also could not leave the building without an employee present.

My talk was about Security Awareness and why we should stop it, or at least have the ambition to make it obsolete. This is obviously a statement to make the audience think about the value of Security Awareness and when and when not to invest in it. When looking to the organization I work for I see that the most value comes from Security Awareness on the level where change is done. Whether it is IT, HR or the Legal department, everywhere there can be made a change there can be made a difference. Obviously the Security Awareness in every department is for the most part completely different.

I ended with my talk with an advice that Security Awareness for IT departments should focus on automation. The more you automate, the more predictable and agile you will become. And when you are agile, you can even become anti-fragile. Every time an IT-department consider training users on Security, we should first ask ourselves if we can make our technology better. If not, then we need to question if we can make our policies, procedures and baselines better. And then, and only then, we can start training users. Because leaning on awareness for security, is leaning on the weakest link in the chain of security, the humans.

And again, for Security Awareness in general, focus it at the places where changes are done in order to really make a difference!

If you want to read more about my point-of-view concerning awareness, read these posts of mine.
I really want to share my gratitude towards ION-IP, WhiteHat Security and the United States Consulate in Amsterdam for giving me the opportunity to talk at the event and help creating awareness within the field of Cyber Security and Agile Development.

If you have questions or want to debate or challenge my point-of-view! Please do so! Sharing opinions is creating knowledge and knowledge leads to wisdom! So feel free to comment below.

Sunday, June 12, 2016

Awareness and Cyber Security versus Agile Development

Let me start with a message (yet again): "If you want an agile environment, you need Security Awareness". What did I just say? I recently said in another blogpost of mine that if you need Security Awareness to be secure, then you are doomed in the first place. I still think that is true, but it is only a part of the case I want to make. In this post you will read the rest of my point of view on this topic.

On June the 14th of 2016 I will talk about this specific topic on the “Cyber Security v Agile eCommerce”, hosted by ION-IP, WhiteHat Security & The Embassy of the United States of America in Amsterdam. In this post I will dig a bit deeper on this topic, why I have this opinion and why I want to share it with you and the rest of the world (the world, because if I may believe Google Analytics, my readers are from everywhere, thank you all!).

Security awareness by itself is by no means the holy grail to a secure environment to work, play, and live in. Though many companies and security professionals alike often focus heavily on security awareness, and to be more precise, security awareness focused at user level. I believe that when you cannot make security ubiquitous by nature, you will fail for sure at creating awareness at that level.

A couple of months ago I was looking for a replacement car (my former broke down) and as I really do not know anything about cars, other than how to drive one at least, I noticed a certain behavior of mine. And that behavior is commonly known as “a typical user who does not know anything technical about the topic at hand and therefore lacks any sensible judgement about its mechanics”. And it gave me a valuable insight concerning security awareness at user level.

This insight was the very fact that I literally did not ask one question about the security of the car. I did not ask anything about the airbags, brakes, seat-belts, electronic brake-force distribution, lane detection, and the security features accompanied with cruise control. I asked myself why as me being a security officer, and the only answer that pops up is this: “If I do not know anything technical about a car, I for sure cannot influence its security, I can only use it as it is build”. I think that the security of the car should ubiquitous. And this changed my perspective on creating security awareness at user level.

The point that I am making in this talk is that your focus or drive should be to eliminate security awareness at user level. This focus helps you to become more creative by building solutions so non-technical users are working secure by design, instead of working insecure and hopefully with some awareness stuff will not hit the fan. But this needs security awareness at another level in the organization.

My statement in the talk is this: “Move security awareness to the level where change is done…”. If the people with the power to change are aware of security, chances are more likely that a securely designed system is being created. So instead of focusing all that energy, time, and money on security awareness for everyone in the organization, try focusing it all on your IT staff. I firmly believe that when you do that, nice things start happening. Part of this firm believe is the experiences I had with this principle in the company where I am fortunate enough to work at.

Security and privacy are not the same thing, so I also advocate to start working on privacy awareness. And, in contrast to security awareness, that topic should be addressed to everyone in the organization. It is by far more effective than security awareness, because the topics are more relevant by nature, and people can really influence them by their behavior. It is easier to teach a user to not share a social security numbers or medical details by phone and e-mail, than learn such user to recognize specially targeted and crafted phishing mail and to not click on the link.
Source: VISTA infosec
Let me end my post with an advice and a question.

When you focus security awareness on the place where change is done, you get more (and this will grow over time) systems designed securely. Awareness should also focus on automation. Because I firmly believe automation is imperative for an increased security. And the more you automate, the more agile you will become. And the more agile you are, the more anti-fragile you can become. And anti-fragility helps making your business become weatherproof for (sudden) changes in its environment.

So, Security awareness at IT-level leads to secure design, which leads to automation, which leads to increased agility which can lead to anti-fragility. Cyber Security can go hand in hand with Agile Development. And let me turn that around, a strong Agile Development culture can greatly increase your security!

What can you design and automate in order to make security awareness obsolete, while increasing the agility of your business?