Cyber Catalyst; Dead Cert or Rank Outsider?


Disclaimer: The views in this article/blog posting are my own opinion based on the available data that Marsh has made public.

As mentioned in episode 3 of my OMG Cyber! podcast, a number of insurers/brokers have joined a new cyber ratings project known as “Cyber Catalyst”.

More details can be found here: https://www.darkreading.com/risk/insurers-collaborate-on-cybersecurity-ratings/d/d-id/1334258 and direct from Marsh here: https://www.marsh.com/us/campaigns/cyber-catalyst-by-marsh.html

Here are a few snippets from the article on the Marsh site:

In the Cyber CatalystSM program, leading cyber insurers evaluate and identify solutions they consider effective in reducing cyber risk. Participating insurers include Allianz; AXIS; AXA XL, a division of AXA; Beazley; CFC; Munich Re; Sompo International; and Zurich North America. Microsoft is a technical advisor to the program.

Cybersecurity products and services viewed as effective in reducing cyber risk will be designated as “Cyber CatalystSM”. Organizations that adopt Cyber Catalyst-designated solutions may qualify for enhanced terms and conditions on cyber insurance policies from participating insurers.

I applaud Marsh for doing something to try and address the lack of cyber risk analysis, profiling, etc. However, I do question the value of this initiative; I will outline below my concerns and thoughts on why this is, I believe, not a helpful offering.

I do not see the value of Insurers/brokers carrying out product/solution ratings, as:

1. They (the insurers/brokers) are not experts in this area, and
2. There are already plenty of other independent testing/rating organisations that have been doing this for many years, to a very high standard. These include ISCA, NIST, AV-Test, and so on… It would have been far more sensible to partner with one of these instead, and it would have added more credibility…

So, this seems to be a strange thing to attempt; a bit like reinventing the wheel and coming up with a different shape that is not as efficient as the one we already have which has served us rather well, so far.

The program is, by my understanding, stating that if a client/insured has product/service x, y or z from the list of “approved/recommended” ones, that the client will get better rates (such as higher limits/lower premiums) and so on.

1. Now, this is fine, apart from the perspective that just because the client/insured has purchased an “approved/recommended” product/solution, it does not mean that they have rolled it out or installed it.
2. Even if they have done so, where are the checks and balances to confirm this, that it is not only rolled out, but actually configured correctly?
3. Furthermore, where is the ongoing validation? Without that, this is pretty much just a box ticking exercise, and therefore no better than the existing risk rating mechanisms they already use.
4. They state that “Microsoft is a technical advisor to the program.”, this does not really help, as they are not a trusted independent review organisation/body. What happens when Microsoft review their own products and solutions?
5. Their disclaimer doesn’t exactly offer a ringing endorsement of the value of the program, read it for yourself and see if you agree?

I would say that this is little more than a “beauty contest” and it doesn’t really do anything to address cyber risk in a new way.

Now, just to be completely transparent, I used to work for AIG as a Cyber Risk Specialist (and so I understand Cyber Insurance quite well). I helped AIG design their Cyber rating solution known as “CyberMatics”. Let me be very clear, I have no axe to grind with any of the insurers, and receive no financial benefit from “CyberMatics” or AIG on this, or any other article/blog posting that covers cyber insurance.

The difference with “CyberMatics” is that is collects telemetry and/or meta data to validate that:

1. The insured has the solution/service installed correctly, and more importantly
2. That it is being used correctly; not just once, but on-going, and this is shared with the client/insured via a secure portal, to help them further improve their cyber defences and resilience.

That is a huge difference!

You can find out more about “CyberMatics” here: https://www.aig.com/business/insurance/cyber-insurance/cybermatics

What are your thoughts on this?  Please let me know…

“Cyber Catalyst” and “Cyber Catalyst by Marsh” are registered trademarks of Marsh LLC
“CyberMatics” is a registered trademark of AIG

Insurance, Silent Cyber, and Refused Claims, Oh My!

This is a companion blog posting to my Episode 1 Podcast about Insurance, etc. which can be found on the Podcast page of this site, or on all good podcasting platforms, including Google, Apple, Spotify, Pocket Casts, etc.

Disclaimer

I am not an insurance specialist, I am a techie with over 30 years of real-world experience in malware, over 15 years of ethical hacking experience and over 10 years of digital forensics (incident response) as well as working for a large cyber insurer for over 2 years (note past tense) where I worked hand-in-glove with underwriters, brokers and claims staff in helping them understand cyber risks, defences and remediation. I also used to meet with CISOs, IT Security Managers and Risk Managers/Legal Council to understand their risks and processes, procedures, technologies, business partners, supply chain and cloud/outsourced services.

I run my own business; I do not work for an insurer or sell insurance (of any type). However, when I did work for an insurer, along with being the cyber risk specialist assisting underwriters, brokers and claims adjusters. I also trained many cyber underwriters, helping them to understand the technology, the lingo (acronyms) and what are the right questions to ask (and what are good answers), when to ask them, and to who (so that they could have meaningful risk dialogue with CISOs, IT Managers, etc.) The underwriters then can understand the answers given and price the risk appropriately, rather than just fearing a worse case scenario, and pricing according to their fears/expectations (which is far better situation (both on cover/limits and pricing) for the insured/client too)!

“Silent” Cyber

For those of you that are not in the insurance industry, you may not be aware of this term and what the implications are to existing (non-Cyber) policies, such as Property, Casualty, D&O, Kidnap and Ransom or Crime.

In simple terms, Silent Cyber is used to describe the case where cover for Cyber threats is not explicitly mentioned in the policy wording/coverage. As the insurers would say, these non-Cyber policies do not have “affirmative” cover.

What this means to you as a policy holder is that the insurer may not honour a claim if it is Cyber related for a non-Cyber policy (even if you have a Cyber extension to that non-Cyber policy). Why, because the wording and terms and conditions in force will be those from the master policy (the non-Cyber one/the main policy). This can cause claims to be rejected, as can be seen in the next section of this article.

Refused Claims

There have been two recent cases reported where the insurer has declined to pay a claim in relation to the NotPetya attacks back in June 2017, these are Hiscox vs DLA Piper and Modelez vs Zurich.

Despite what the press and other media has claimed, in both cases that the policy was a cyber policy and the reason stated by the press or other media for the claim being declined was down to an “act of war, or hostile action”.

From what I have found out, neither of these claims are in relation to Cyber Insurance policies, in fact they both are related to Property policies (which are, even with a cyber extension added on, not the same as a dedicated Cyber Policy.   Very sloppy reporting, which doesn’t help anyone…

So, this has resulted in every person and their pet of choice making statements, such as “well, what is the point of buying insurance as the insurer will weasle their way out of having to pay” and “there is no point in buying cyber insurance, as I’ve seen what happened to the claims from Mondelez and DLA Piper”.

Expecting wide-ranging/expert Cyber coverage from a Property policy is like expecting wide-ranging/expert Health insurance from your House and Contents policy! Not surprisingly you will not get comprehensive health cover backed by experts in this area. It’s a bit like expecting your gardener to offer health screening (without them being a medical practitioner).

A few days a go a written statement was sent to SC Media UK (owner of the SC Magazine) in which Kylie O’Connor, the head of group communications at Hiscox stated “The dispute we are in with DLA Piper, is not about a cyber policy and has nothing to do with a war exclusion.” This just proves that the press and other media were (shock, horror) making things up so that they could publish (without little things like “facts” get in the way!

However, in the case of Norsk Hydro, they do have a dedicated Cyber policy, and therefore are covered under that policy (up to their limit, and after taking into account any excess, waiting period, and loss adjustment).

Why do companies invest in cyber insurance?

Well, for lots of reasons, including the ones listed below:

  • Hacking (external or internal misuse)
  • Physical loss of data (left on train, back of cab, accidents (sending data to the wrong person, etc.)
  • Data corruption or eraser (cost to recover or recreate), even paying for ransomware decryption keys.
  • Business Interruption, such as DDoS, Ransomware, etc. including loss of business
  • Costs for first response (forensics, legal, PR), etc. covered under the policy
  • PCI and other fines covered (where legally allowed)
  • Bricking (where a device becomes unusable due to a firmware or other update failing).
  • Legal or contractual requirements (from industry, business partners, etc.)
  • In some cases the insurer will offer services/solutions/products to help the insured improve their overall security posture/maturity for free (as part of the policy) or at a discounted price.

At the end of the day suffering a cyber breach has almost become “normal” and “expected” as not a day seems to go by when we don’t hear about yet another breach (new or historical); a good cyber insurance policy can help offset the risk and related costs for such breaches/incidents.

Then there are new risks/attacks such as CryptoJacking and Password Spraying (O365 and GSuite targeted via IMAP and even if 2FA or MFA is enabled they may be able to get in to your account).

What are  the ways that companies could avoid falling into this crevasse?

Check the policy you have is fit for purpose, check with your insurer or broker. I strongly suggest that you ask your insurer or broker which scenarios/risks you are covered for by the policy and if you identify gaps in your existing coverage decide if the cost of taking out extra insurance is a good risk/benefit trade-off or solution.

Check that the coverage includes first response (forensics, legal and PR services), that you have enough cover for business interruption, including lost business and remediation costs. Also consider the brand/reputational damage and knock-on customer effects, loss of trust, etc.

Check to see that the policy will cover financial fraud, such as BEC/Fake CEO, employee fraud, if not, find a crime policy that includes this. Crime policies are not the same as a Cyber policy as what they cover is different, or from a different perspective.

Make sure that the Limits, waiting period and excess is suitable for your business needs.

Don’t go for the cheapest, especially if the insurer/broker only ask 5-10 questions and doesn’t sit down with your CISO or IT Manager, etc. to discuss the answers afterwards (very few questions can be answered yes or no; they are usually a bit of both and the answer may vary across a typical organisation), as may the questions that should be asked by the Insurer or Broker.

The Future?

Even though a dedicated Cyber policy is a far better bet in today’s incident/breach strewn world, there are some things that they still don’t cover.

I want to see the Insurance industry step up and make Cyber policies more inclusive; it would be better if Crime cover was also included (including not only crime and fraud due to hacking, but also fraud due to social engineering or insiders/insider collusion). This should include BEC/Fake CEO and Invoices, etc. even when NO hacking or breach has occurred!

In Summary

Organisations need to ensure (no pun intended) that the existing Insurance policy or policies they have are fit for purpose and will actually pay-out when needed. You need to purchase the right policy type for the right risk, as otherwise you could end up in the same situation as DLA Piper and Mondelez… If in doubt check with your insurer or broker, before it is too late!

Update 15th April, 2019: It has come to my attention that Merck is also suing their insurer for refusing a claim; again it is NOT a Cyber policy, it is in relation to their Property policy.

What Cyber Threats and Trends Might We See in 2019?

‘Tis the season to get out the crystal ball and play at being the cyber equivalent of “Mystic Meg” (no that’s not me in the picture).

For 2018 I predicted a number of things that were spot on, these included the following:

  • The change from mass ransomware campaigns to more targeted ones asking for higher ransom payments.
  • The move from ransomware to cryptomining/cryptojacking as the primary monetisation payload/method.
  • GDPR being used for extortion/blackmail attempts.
  • Organisations still not focussing on the basics and best practice for their industry/vertical and wondering why they suffered security breaches/incidents.

So what will 2019 bring, according to OMG?

  • More targeted extortion attempts; Ransomware, GDPR, DDoS, etc. All with higher ransom being demanded.
  • Organisations will still be mainly focussed on the latest, must have “shiny toys/technologies” rather than dealing with the basics and best practice for their industry/vertical.
  • A mainstream move towards two or multi-factor authentication, as password theft is increasingly seen as the main way that bad guys and girls get in; other than social engineering (phishing) or via the supply-chain/business-partner. This move will be required due to massive Credential Stuffing attacks in 2018 fuelled by the many data breaches where user ids and passwords were stolen.
  • More supply-chain breaches as a method to gain access to the intended victim organisation.
  • Cloud service breaches and/or take-downs and mis-use by the Bad Guys n Girls.
  • The skills-gap and staff shortage will increase, again. And those of us in the industry will be in demand and frequently head-hunted or just pestered by desperate recruiters that don’t read your LinkedIn profile and still approach you with roles that you are not interested in or have the skills/background for.
  • More Business Email Compromise attacks (aka Fake CEO/CFO, etc.); these will rake in far more money in 2019
  • Artificial Intelligence and Machine Learning will continue be touted as “The” solution to deal with cyber threats and breaches; they are useful but generally too prone to false positives (detect things that are not an issue) and more worryingly false negatives (don’t detect what they should do).
  • The Internet of Things will start to “grow-up” as manufactures start to bake in security and offer it as a differentiator to competing products/services.
  • However, despite this we will continue to see IoT devices/infrastructure used as an attack platform and I suspect that we will start to see volumetric DDoS attacks exceed 2Tbps (largest so far was 1.35Tbps against Github in 2018). 
  • We may well see some critical infrastructure attacks (outside of Ukraine) that are successful, and that cause major outages and/or physical damage/loss of life.
  • Too many organisation thinking that using a single Cloud provider will give them a fully resilient infrastructure; it won’t. Just like having multiple data-centers, you need multiple Cloud providers (this should be part of your Business Continuity and Disaster Recovery Plan), no single-points of failure!
  • GDPR will finally start to bite (hard) and organisations that should have already been following industry best practice for data/privacy will finally do something about it (well, most of them)!
  • Blockchain will be finally recognised as not being the solution to everything!
  • Increase in use of Sextortion, Bomb and other extortion/blackmail emails/calls, despite the fact that most Sextortion campaigns did not net piles of bitcoin as those behind them expected.
  • More social-media scams mainly focussed around crypto-currency giveaways; like the many Elon Musk themed ones we saw in 2018.
  • People will still mainly fail to learn from history; we will see yet more old techniques/technologies dusted off and re-used by the Bad Guys n Girls, for victims that weren’t around (or paying attention) the last time it was successfully used…

Don’t have nightmares, remember that 80-90% of all security breaches/incident I have dealt with could have been avoided by just following best practice and doing the basics… This includes taking (and testing) backups, educating (and testing) your staff, patching your systems, applications and writing secure code, good Identity and Access Management, and so on…

Effective End-User Training, Compliance and Testing

What do most staff think when they hear the words “end-user security training” or “security awareness training“?

They think, “Oh no, is it really that time of the year again? What a waste of my time; it is so boring and doesn’t teach me anything that is useful to me. Security is the IT department’s problem/job, not mine!

In many organisations security awareness training is dull, impersonal and does not use “real life or real world” stories to add colour and flavour and help explain the problem, the risks, and the impact of security breaches/incidents. In most cases the training is to read a policy document (Internet Usage Policy/Security Policy, etc.) or to watch a video or attend a webinar where they are preached to rather than being allowed to participate in an interactive or interesting session.

Staff need to understand that in today’s world, security is everyone’s job, because if your staff are not part of the solution, they are part (if not most*) of the problem!

[*] 95% of successful cyber attacks are the result of a phishing scam.
Source: (2017) Ironscales,Email Security Report.

Add to this that many staff treat end-point protection (anti-malware, personal firewall and related security tooling) as an “authentication” method; “if I can open this link/file in the email, go to this site, etc. and my system gets hacked/infected, it is not my fault, it is the security/IT departments fault!” This is captured very nicely in this cartoon.

Given the above perceptions of many staff/end-users, what can we do to try and reverse this situation, so that staff see security as part of their job/responsibility and become part of the extended security team?

What do you need to make end-user security training successful, rather than something that is hated/despised/loathed and avoided at all costs for as long as possible by most staff in almost every organisation?

Here are some top tips:

  1. Make it fun; use gamification, where they are engaged, entertained, involved and tested throughout each module.
  2. Keep it short and punchy; no longer than 20 minutes, backed up with bite-size (5 minute) modules to reinforce an individual topic/threat. Don’t try and do the whole organisation at the same time; do it in groups and stagger the roll-out to be more effective.
  3. If you can make it a competition; who can report the most spam/scams/phishing emails, etc. Give prizes, or at least recognition!
  4. Make is personal; teach them skills that they can use in everyday life, including at home.
  5. Phish your own staff (after training them, and before) so that you can gauge the effectiveness of the training, but do it wisely and sparingly as otherwise they will quickly become fatigued and disinterested.
  6. Don’t penalise those that fall for the phishing test emails; use this instead as a “teachable (not preachable) moment“, rather than shame or blame them, try to understand why they fell for it, and explain how they could have recognised it for what it was.
  7. Make sure you set-up an email address such as: “[email protected] which can be used by employees when they suspect they have received a phishing email. Explain what steps they should take in order to report the email and give them with the necessary tools/guidance to report a suspected phishing email, such as a “report-phish” button in their email program.
  8. Training is not a one-time or once a year thing; good awareness training is part of the culture of an organisation and needs to be topped-up and refreshed all the time to stay effective. Make sure all staff, from the C-Suite down to the most junior staff in the organisation are included, not just techies.
  9. Ask for feedback, especially ask them about what they are worried about, e.g. Ransomware, Scams, Sextortion, Social Networks, Privacy, Passwords, GDPR, Data Breaches and how it impacts them personally and the company/organisation/industry, etc.
  10. You could always bring in a real-life “hacker” (an Ethical one, also known as a Penetration Tester or White Hat Hacker) and let them talk to your staff and answer their questions; they will have lots of real-world stories and good advice. T here are some that are good at talking to non-techies without resorting to acronyms and technical jargon.  These rare individuals will use humour, analogies and stories to help illustrate and bring the subject to life; they will often be very passionate about security, and this will keep the audience engaged.

However, you will find that 10-20% of your staff will just not be trainable (from a security awareness perspective)and you need to identify them and work on ways to reduce the risk that they pose to your organisation.

As the old saying goes:
The Bad Guys n Girls only have to get lucky once;
the Good Guys n Girls have to be lucky all the time
“.

So, what is a good solution that isn’t going to break your budget, but still allow you to deliver most of the above as a managed service and tie in to your Active Directory do that you can assign training to groups or individuals and see the results (meta data) from the training and testing?

One vendor that I have found to be very effective in this space is Techguard Security, this is what they say about their offering:

“Empowering your workforce to recognize and respond to sophisticated threats is only a click away. TechGuard S.H.I.E.L.D is a cutting-edge and comprehensive training solution for businesses of all sizes.”

You can use the following link to find out more about Techguard and their offerings, including the end-user training and phishing testing offerings, and what’s more, if you decide you like what you see and sign-up with them, you will get 10% off the price!

To find out more and claim your 10% discount when you sign up, use this unique web link: https://www.techguard.com/omg-cyber-security/

If you don’t use that link to register your interest, you won’t get the discount when you sign up.

Don’t just take my word for the effectiveness of good Security Awareness training, here are some statistics:

  • According to research by Ponemon, even the least effective training programmes have a 7-fold return on investment.
  • Most cybersecurity training programmes result in a 37-fold return on investment.
    Source: (2015) Maria Korolov, Does security awareness training even work?

If you don’t train your staff and carry out phishing tests, the Bad Guys n Girls will, and the results won’t be pretty…

I have been doing security for over 30 years and I often state “The day I stop learning will be the day they bury me“, in other-words, I’m still learning and will continue to do so until I die.