CISSP Domain 3: What FIPS 140-2 Going Historical Actually Means (370)

Machine-generated from the episode audio, with names and acronyms corrected. Spoken wording is otherwise unedited, so expect the occasional rough sentence.

Welcome And Domain 3 Focus 0:00

Hey, I'll Shon Gerber with CISSP Cyber Training and hope you all are having a beautifully blessed day today. Today is Monday, and we are gonna be getting into the CISSP exam, or aspects related to domain three of the exit the CISSP exam. So we're gonna be getting into some aspects related to cryptography. And of that domain three, it's 3.3 and 3.4. So we're gonna kind of give you just a little bit of an analogy here as we get started, and then we'll kind of roll into some of the concepts around this because cryptography can get a little bit wonky, and it's not because it's not it's hard, it's just the fact that it's just a different and it's a different wording and a different language we kind of use in the cryptographic world.

So we're just gonna kind of go with a little bit of analogy and then from there we'll we'll dig into some of the content. So

The Sticker Analogy For Compliance 1:12

picture this you have a safety inspection sticker on your windshield. A mechanic has checked the car, signed it off, gave you the sticker with a date on it, says you're good to go, right? So now imagine the state, whatever state you live in, announces it's switching to a new inspection standard. And the one on that specific morning, every sticker issued under the old standard stops being accepted. So your brakes did not get worse overnight, the car is exactly as safe as it was yesterday. What changed that you can no longer what can happen? So what can you do? What are you how are you gonna deal with this situation?

So that happens to a huge amount of the world's encryption a week from now. And nobody's affected by it, nobody understands what's gonna be affected by it, and they don't have this date on their calendar. So nobody gets hacked, no data moves, no alarm fires. Basically, this is what happens. A list gets updated, and that is it. And this is what we're dealing with with domain three of the CISSP exam. So now that we've given you this a little bit of this analogy, let's kind of walk through this a little bit and kind of go down this path.

What A Cryptographic Module Is 2:12

So when your software encrypts something, it doesn't do that itself. It hands the job to a specific sealed-off piece of code that does the math and holds the keys. That piece has a name. It is called a module. So back to the car analogy, your product is the car. The module is the part that gets inspected, a e the brakes, or I. e. Now the government runs the inspection program for those parts. A laboratory tests the module against the published standard, and then it passes. It gets listed with a certificate number that anybody can look up. So the standard is called PHPs, and we've talked about this a lot on CISSP Cyber Training, the FIPS 140 series.

And the certificate is called a validation. So that's your sticker. The validation is your sticker. So on September 21st, the old standard stickers stop being accepted. Every remaining certificate under the previous version moves to what they call historical status. So let me be precise, because a lot of coverage is getting this wrong. Nothing is revoked, nothing stops working, your VPN still encrypts, your database still encrypts, the math is unchanged, your breaks are fine. And again, there's nothing wrong with this. It's just a change in the compliance aspects of this. But they're doing this, right? Because there is a path forward. So the product keeps working, it just stops counting.

So why this reaches people who work for the government, right? Because the FIPS validation got written into an enormous number of the contracts that have nothing to do with the federal government. The clearest example is defense contracting. If a company handles sensitive government information, the kind that isn't classified, but you still can't leave it lying around, its contract requires validated encryption. And we use this term a lot. And it's you'll be finding this out very interesting as it relates to the CISSP and just security in general. People throw out validated encryption a lot and they really don't know what they're talking about. So we're not talking about strong encryption.

We're not talking that encryption that was switched on incorrectly or any of those things. They're just talking validated. And the enforcement tightens November. So now to be fair, nobody has banned old certificates. There's no prohibition on anything. But picture the assessor that they come in, they open your folder of evidence, and they find the sticker that expired three months ago. That's the point. So the best case, they ask you, what is your plan? Worst case, they write it up. Neither one is a conversation you want to be having for the first time in the room. Again, we do we talk about this with auditors. You want to make sure that you have your stuff buttoned up and tight and at least, at a minimum, have the ability to discuss where you have challenges, right?

So if you see this FIPS issue that we're gonna be talking about, turn on, come on, how do you deal with it?

FIPS Mode Versus FIPS Validation 4:51

The distinction of this whole process, this is how we deal with it. Here it is, you write this down. Running in the FIPS mode is not the same thing as holding a FIPS validation. So again, running in FIPS mode is not the same thing as holding a FIPS validation. FIPS mode is a switch. You flip a setting. The software limits itself to improved encryption and a green check appears somewhere on the screen. Depends on what kind of client software you're using, but that's what it does. In the car, that's telling you the trooper, basically your brakes are fine. My brakes are good, no problem. The validation is the sticker.

Somebody else tested it, signed it, and put it a number on it that anyone can go verify. So one is something you say about yourself, the other is evidence a third party is produced. And here's why it matters far past this deadline. So if you're asked, is your encryption validated? A lot of organizations will honestly answer, our encryption is configured. Both are true. In a status meeting, they sound identical. In an assessment, they're not the same. They are not. So let's kind of focus on this and see where we're at.

Historical Status And 500 Day Queues 5:51

Now, the part that turns this from paperwork into a real problem is there is a new standard. And getting a module tested and test listed under it currently takes, on average, more than 500 days. Basically around almost two years, right? So one inspection station booked out 18 months. That's what it's saying. So you can't get it validated for 18 months. We're talking about the car analogy again. And another government deadline says certain new purchases have to meet an even newer encryption suite by January 2027. Those need their own testing as well. So do the arithmetic out loud. A vendor getting in line today does not come out of the other end before that date, right?

We're talking just a few months away. So for products only starting now, the timeline doesn't work. Not because nobody's lazy, but because the line is longer than the runway. Some products are already through and fine. The problem is everyone is still cued. And that's exactly the position the CISSP wants to see how you handle. When a requirement generally cannot be met in a time available, a manager doesn't pretend otherwise and doesn't quietly fail. You write the gap down, you treat it as a risk, and then get somebody with authority to accept it or do so on the record, right? A known gap with a decision attached is a managed risk.

The same gap nobody wrote down is a finding. So why all these deadlines exist? The

Harvest Now Decrypt Later Explained 7:12

last thing that I want to kind of bring up here is what makes this urgent instead of bureaucratic. Nobody has a quantum computer that can break today's encryption. Not today. Not this year, probably not next year. Anyone telling you otherwise is selling you something else. The actual threat is this somebody copies your encrypted traffic right now. Can't read it on the word of it. They just basically copy it, they store it, and they wait for the machine that can open it 10, 20, who knows how many years from now, but that is what the case is going to be. Now it has a name. It's harvest now, decrypt later.

So it flips something most people assume without thinking about it. You normally picture encryption as protecting data until you stop caring about it. The real question is how long your data has to stay secret. Measured against how long your encryption has left. If those two numbers overlap, you already have a problem. You just can't see it yet. That's the challenge. That's the line I want you carrying into the exam room, right? The encryption protects your data as long as the algorithm holds, not for as long as you need it. And we've talked about this over and over again, but it's the truth. It truly is.

Study Programs And Resources Plug 8:18

Okay, so quickly before we roll on, I gotta do a shameless plug for CISSP cyber training. So if you're studying a lot but not retaining a lot, that's a method problem, not a discipline problem. That's what happened to me the first time, and that's why I failed. Accelerator is $19 a month. It is a question bank and domain walkthroughs. Pro is $39 a month and adds live office hours, CPE insider sheets with every episode and documented CPE each week, pre-filled and ready to go. And the sprint cohort is for people who really want to get this done in the next two months and just want to just dive in as back as best they can.

They have a test date, they want to put it in place, and they want to be ready by it. So go check it out. CISSP Cyber Training. com. All of those things are available for you and check them out. All right, so let's move on to our training.

Cryptography As A Managed Life Cycle 9:07

Okay, so we are now going to get into cryptographic solutions. This is all for domain CISSP domain 3, 3.6, and 3.7. This is going to be dealing with lifecycle, validation, key management, PKI, cryptanalysis, and post quantum. These are all big factors that you will see on the exam in some form or fashion, and we are here to show you how they work. So this material candidates are probably the most afraid of, and honestly, the material exam is the gentlest about. It does not want you to do the math, it wants you to make decisions about cryptographic cryptography like a manager. You select it, you manage it, you retire it, and you know what to do and what it does to protect it.

That's the overall goal. They don't want to get into the gory details because honestly, it's very challenging for many, many people. But they want you to understand it and think like a manager and how would a manager deal with it. So start here because everything else hangs off of it. Cryptography is a life cycle, and the exam tests it as a life cycle. You select an algorithm and a key length, and based on what you're protecting and for how long. You implement it in a module with a configuration. You operate it, which mostly means you're managing the keys. You assess it on a schedule against what has changed in the world, which we talked about previously.

And then you retire it deliberately, before it fails, not after, right? The last phase is where organizations fall down, and it is almost always this is the area where many people and organizations will fall down. And it's this week where this week's deadlines actually make a big difference. Almost nobody plans for the retirement of an algorithm at the moment they adopt it. The CISSP expects you to do that. The term for us doing this well is called cryptographic agility. Building systems so that an algorithm can be swapped out without re-architecting the application. If your crypto is hard coded, you don't have a migration project. You have a rewrite.

And we talk about this with the developers all the time. Do not hard code crypto or keys into your development code. This slide is the nuts and bolts of what we're talking about today in this episode. There are three different claims people make about cryptography, and candidates routinely treat them as one. The first, we use strong encryption. That's an algorithm claim. AES 256, it has nothing about how it was implemented, right? So your AES 256 is what I'm deploying. That is an algorithm claim. The second is we run in FIPS mode. That is a configuration claim. A flag is set and a software is restricting itself to approved algorithms based on the FIPS monitoring that is built into it.

So basically, they're not the same thing. The third, we hold a FIPS validation. That's an evidence claim. The named version of a named module is submitted to an accredited laboratory, tested against the standard, and issued a certificate number by the cryptographic module validation program. They're otherwise known as the CMVP. Notice what separates them. The first two are things you say about yourself. The third is the things somebody else says to verify and verified about you and is published. The exam cares about the difference enormously, and so does an auditor. So here's the heuristic around this. When a question offers you an answer that sounds like a security posture, ask who verified it?

And where would I go to look it up? If there is no answer, it's an assertion. It's somebody just making a statement.

FIPS 140-3 Changes Worth Knowing 12:27

Okay, so the FIPS 140-2 versus 140-3. So you don't need to memorize this standard. You need to know what kind of thing it actually is. FIPS 140 validates cryptographic modules, the boundary around the crypto, not the algorithm inside. The algorithms are approved separately. 140-2 was a self-contained American standard. 140-3 is a different in a way worth knowing. It's essentially a wrapper around two international standards. ISO 197 19,790 for the security requirements, and ISO 24,759 for the test requirements, with the US specific modifications published in NIST's 800-140 series. So a few changes that show up in questions. 140-3 adds an umbrella term. And understand this because it's exactly the kind of vocabulary the exam tests.

Sensitive security parameters, SSPs, cover two things. Critical security parameters, your secret and your private keys, which need confidentiality and integrity. And the second thing is public security parameters, public keys and certificates, which need integrity only. The CSP did not go away. It got a sibling. And that's nice one because it corrects a common misconception. Public keys are public, but their integrity is still a security property. The point you gotta think about is because public keys are in the public, that doesn't mean that they're just they can be anything. They are still security related. Power on self-tests become pre-operational self-tests, so integrity checks run before the module offers any service.

And on the flip side, conditional algorithms self-test now can wait until the algorithm is actually used rather than firing it all up at startup. So the 140-3 permits a degraded mode, a module that can keep offering reduced service after a failure instead of shutting down entirely. There are four security levels. Level one, approved algorithms, production grade, no physical authorization required at all. Level two adds tamper evidence and role-based authentication. Level three adds tamper detection and response, plus identity-based authentication. Level four is a full protective envelope with environmental failure protection, and it now requires multi-factor identity-based authentication. So there are four specific levels that you will have to potentially be aware of.

The exam framing, the FIPS 140, is about assurance. It's the answer. How do you know, not what algorithm did you pick? Okay, so a point that we need to really drive home here is almost nobody can break

Key Management Where Crypto Fails 15:06

the algorithm, right? That's not it. They get the key. So when somebody's trying to get access to something, it's not the fact that they broke the algorithm, they actually got the key. Key management has its own life cycle and the exam tests every stage of this. And I would say this is one of the biggest things you need to truly understand is key management, both from the test standpoint and from the fact that you know what, you're going to be dealing with this in your jobs in various ways. So key management has its own life cycle and the exam tests every stage. Key management is where the cryptography actually fails.

It just does. So very rarely is an algorithm actually broken. It's only really broken if somebody uses an algorithm that hasn't been tested. They get the key. And key management has its own life cycle, and the exam tests every stage of this. And one of the factors you're going to need to be aware of is this is probably the areas where most companies do fall down. So with a proper source of randomness, there's generational aspects. Weak entry of a generation poisons everything downstream, and no amount of key length will fix it. Distribution. This is getting the key to the other party without exposing it. This is the problem asymmetric cryptography has was invented to basically to solve.

Storage. Ideally, the hardware security module is where you want to put it. So the key never exists in plain text in general purpose memory. It's all through a storage in a hardware security module. Use with separation of duties, so no one can one person can both use a key and change who else can. Rotation on a schedule and on compromise. Escrow and recovery, so lost keys don't mean lost data with dual control, so escrow does not become a backdoor. And then destruction. Deliberate, verified, and remember this for domain two. Destroying a key is a recognized way to make the ciphertext unrecoverable. Cryptographic erase. We talked about that in domain two last couple weeks ago.

The key piece on this is understanding the destruction phase in this in this plan. The exam's favorite key management traps are all about split control. Dual control means two people are required to perform one action. Split knowledge means no single person has the entire key. Those three get swapped in distractors continuously all the time.

PKI Trust Expiration And Revocation 17:28

Okay, PKI, public key infrastructure. This is something that the trust plumbing that's under everything out there. A certificate authority vouches for the binding between the identity and the public key. A registration authority verifies the identity before it's actually issued. And then the CA signs it, the certificate authority. The revocation gets tested. A certificate revocation list is a periodic published list of revoked serial numbers, simple and stale between the publications. And they'll bring up concepts around the revocation list on the CISSP exam. OCSP checks a single certificate in real time, current, and it tells a responder what you're visiting. OCSP stapling fixes that privacy problem by having the server present a recent signed status itself.

And the concept that ties PKI to the rest of the domain is a certificate is a binding with every expiration date. You trust in a PKI is deliberately timed box. It is not an inconvenience. It is the overall design. Anything that grants trust forever is broken by construction. Do not, again, self-signed search. All of those things. When it's trust forever,

Cryptanalysis Categories And Assumptions 18:40

it's not good. Okay, so we're gonna get into different types of attacks and the aspects related to those. So domain 3.7, know the categories and more importantly, know what each one assumes the attacker has. So brute force, try every key. This is defeated by key length. Cipher text only, the attacker has only ciphertext. The hardest case for them to debate to understand. Known plain text, they have the matching plane and cipher pairs. Chosen plain text, they can make you encrypt the text they picked, and the chosen ciphertext is the mirror image. Frequency analysis, this is statistical, and it wrecks classical substitution ciphers. And then the category that matters the most in the real world, implementation attacks.

These do not attack the math at all. Side channel attacks read something leaking from the physical system. Timing, power consumption, electromagnetic emissions, and even sound. Timing attacks specifically measure how long an operation takes and infer key material from the variation. Fault injection deliberately induces an error, voltage such as a voltage glitch, heat, radiation, and then reads the secret secrets out of the broken behavior. Here's the framing to carry. The algorithm is almost never the weak point. The implementation, the key management, and this us humans usually are. Which is exactly why FIPS 140 validates the module. The module is where the implementation attack

Quantum Risk And What Breaks 20:02

lands. Okay, so this quantum threat, precisely. So you need to be careful in this space because there's a lot of nonsense out there about this. And we're gonna go into two different algorithms and just kind of briefly touch on those. But the key thing to keep in mind here is that nobody is breaking RSA today. What is known is that the algorithms are vulnerable and why? So we're gonna get into two algorithms. One is Shore's algorithm. On a sufficiency large quantum computer, efficiency solves integer factorization and discrete logarithms. So this is the mathematical foundations of RSA, Diffie-Hellman, and elliptical curve. Those are broken, not weakened.

That's what he's saying. Grover's algorithm gives the quadratic speedup of unstructured search. So quadratic, not exponential. The effect is to have the effective security in bits. So an AES 256 gives you about 128 bits against a quantum attacker. So that's bad. That's weakened it up quite a bit. Right now it's still fine, but in the future, that is going to be a bit challenging. So this is where the HarvestNow decrypt decrypt later is a threat model that makes it urgent today. So an adversary captures encryption now and stores it until the machine exists. That is where they can start causing problems. So this is which produces the only quantum risk calculation that you need.

How long must this data stay confidential? You really need to understand that. Plus, how long will it take for me to migrate versus how long until the capability exists? So if it's confidential and you need to migrate it, is the capability gonna be in five years? Well, you're probably okay. If it's not gonna be, if it's gonna be, or say it's gonna be in one year, you then we have to may have to be worried about the data. So again, you need to understand what is the algorithm that you're using, and then what is the what is the length of all this harvest now decrypt later gonna play out for you?

So if the first two add up to more than a third, so we're saying is that you must know how long the data stays confidential plus how long it will take me to migrate versus how long versus Of the capability list exists, then if it goes up to a third, you're already exposed. So you you need to keep that in mind. Realistically, if you're using an encryption algorithm that's less than 256, um, you probably are exposed considering these two things have come out. Now, if you are if it's 256 and above, you're getting probably push in the envelope, right? That's the part that they're they're bringing

NIST Post Quantum Standards And Dates 22:22

up in this case. So NIST published the first post-quantum standards in August of 2024. Three of them. They have the FIPS 2003 or 203, FIPS 204, and FIPS 205. So 203 is a module latest key encapsulation, and that's the one that replaces the key exchange. It came out of the algorithm formerly called Kyber. FIPS 2020, this is the primary digital signature standard, formerly Dithylinium. FIPS 205 is a stateless hash-based signature scheme, formerly Sphinx Sphinx Plus. It exists as a backup that rests on completely different math. So breaking the latest does not take everything down with it. So notice the pattern of that trio because it's a design principle, it's not trivia.

They deliberately standardize a backup based on unrelated mathematics. Diversity of assumption is a control. So to break this down into some level of understanding around dates, you need to be aware of January 1st, 2027. And it basically means you need to migrate to these new systems or this new standards by January 1st, 2027. Doesn't mean anything's broken, it just means you need to migrate to them. There's a draft IR of 8547 with NIST, and this proposes deprecating RSA 2048. And there is a draft IR 8547 that's dealing with NIST. They're basically talking about how you're going to roll these different algorithms and what you're going to do with quantum vulnerable public key algorithms.

This doesn't come into effect until 2030, 2035. These aren't going to be known for the test. The important part to think about, though, is that they're known for your job. You're going to need to understand these different dates. So the key dates to keep in mind, January 1st, 2027, you need to be putting everything that is in place. You need to maintain and manage their all new acquisitions, need to be required to be compliant by January 1st, 2027.

Inventory First Then Migrate 24:12

Okay, so here is where people get this wrong on the in practice and on the exam. So as to prepare for a post-quantum migration, the practitioner instinct is to start swapping algorithms. You cannot don't do that yet, right? The first step is a cryptographic inventory. We talk about inventory all the time, sometimes called a cryptographic bill of materials, or like we talked about, software bombs, right? Software build of materials. What algorithms are used, where they are, what products are they in, holding what data and what key links, and who owns each one of them. Again, these bombs. The U. S.' s Office of Management and Budget directed federal agencies to inventory their cryptographic systems back in November of 2022.

And they said that they had to have this done because you cannot migrate, which you don't know is out there. So it sounds familiar. It should. It's the same sequence as domain two. You cannot protect what you do not identify. So after an inventory, classify by data lifetime. That's the mox masaka calculation we talked about earlier, the one-third piece of this. Then prioritize what is the longest confidentiality requirement. Then assess your crypto agility because the systems that cannot swap algorithms are your long poles in this tent. And then begin migrating. But really understanding where are all your algorithms and which ones can be swapped out and which ones cannot is a huge factor.

I would, if anything, that's an important piece that you really need to nail down. So here's the order. You inventory them. You then prioritize by data lifetime. Then you assess the agility, you then migrate, and that's it. So there's five takeaways in this space. One, cryptography has a life cycle and retirement is a phase you plan at adoption, not on a crisis after you've discovered it. Two, validation is evidence. Configuration is an assertion. Know which one of these questions it is offering you. And then three, nobody breaks the algorithm. They take the key or they attack the implementation. That's why FIPS 140 validates the module.

Four, quantum breaks factoring, and discrete logs. It dents symmetric. Shores versus Grovers. And remember the replacements are the asymmetric as well. And then five, harvest now, decrypt later means your exposure is set by how long your data must stay secret, not by how strong your cipher is today. And the understanding around this week's deadline is the fact that the organization can be genuinely well encrypted and still fail an assessment. Because encryption and assurance are two different things. One protects the data, the other proves it. Okay, so let's get into some CISSP questions.

CISSP Question Traps And Manager Moves 26:39

Okay, here are four CISSP questions that we're gonna go over. And we're not gonna teach you the answer, we're gonna teach you the trap. So, first scenario a hospital system is preparing for a customer security assessment covering protected data. The database encryption appliance is configured in the FIPS mode and restricted to approved algorithms. The security team has screenshots confirming the setting. The appliance's cryptography module holds a FIPS 140-2 certificate, which moved to historical status on the CMVP list in September. The vendor's FIPS 140-3 submission is still in the validation queue, like it is with a lot of them. The assessor asks whether the module holds a current active FIPS validation.

What is the accurate answer? A yes. Historical status applies only to new federal procurement. The module stays on the active CMVP list for assessments. Yes, that's B. The appliance is operating in FIPS mode using only approved algorithms. C, yes, the appliance uses AES 256, which exceeds the strength of the control requirements. Or D, no, the validation of historical rather than active. And add in FIPS mode is configuration, not a validation. Okay, so which answer is it? Again, we're talking about what is a trap here and what could they be doing when it comes to this aspect? And the answer is D, no, the validation is historical rather than active.

The FIPS mode is a configuration, not a validation. So we talked about this, right? There's three different claims. Get treated as one. Only one of the third is evidence. We use AES 256 is an algorithm claim. We run in FIPS mode is a configuration claim. So you need to understand the differences in that specific space. So the question asks for a current active validation. Historical status moves the module off the active list because it is 140-2 for every other purpose, not just procurement. So because of that, and is waiting in the queue, it is historical. So why the others are wrong? Configuration, not validation. FIPS mode restricts the software to approved algorithms.

A posture you assert, right? Validation is a tested listed certificate. The control asks for the second one. Half true and the half that matters is false. Historically, genuine is not revoked. Nothing was withdrawn from the product and it still works. But historical status is not scoped to procurement. Module comes off the active list. That's it. No more. Strength is not assurance. Algorithm strength and module assurance are separate access. They're separate pieces to this. The FIPS 140 validates the module, the boundary around the crypto, not the algorithm inside. Exceeding the length code does not substitute for a certificate. Okay, next question. A researcher organization transmits clinical trial data over TLS using ECDHE key exchange on the curve, P-256, AES 256 G C M.

The data carries 24, 25-year confidentiality requirements. The organization's threat model explicitly includes a well-sourced nation-state adversarial capability of capturing and storing encrypted traffic in bulk. Which change would most reduce the long-term confidentiality risk? Okay, so we got a couple things going on in here. We got a lot of big terms, we got a lot of big acronyms to maybe go, oh, what's going on? The data carries a 25-year confidentiality requirement. That's a big thing to think about. The organization's threat model explicitly includes a well-resourced nation-state adversary. So that's what they're saying. That's the threat. So what change would most reduce the long-term confidentiality risk? A.

Shorten the TLS session key rotation interval. B. Migrate key establishment to a post-quantum or hybrid mechanism such as MLKEM. C. Move the server's private keys to a hardware security module. Or D increase the elliptic curve from P56 to P521. Okay, so we're looking at this. What is the one that you feel is the correct answer? And the answer is B migrate key establishment to a post-quantum or hybrid mechanism such as M-L-K-E-M. Okay, so the STEM gives you the term MASA's inequality, 25-year confidentiality, and we know the adversary captures and stores traffic because it's a nation state. So one of the things we'll kind of break in here why some of these are wrong, why they're not right.

So D, scaling a defense to the threat bypasses. Okay, so we talk about the overall encryption piece of this. To go from a P56 to a P521 for more security, this is the right instinct against a classical attacker, but the answer is it's the answer most practitioners will give. But again, Shore's algorithm does not brute force the curve, it solves the underlying discrete logarithmic problem and solves for P512 or 521 as well. So you have to make the wall taller against an adversary who's walking around it. That's the point. A rotation does not help against stored cipher. Faster rotation limits basically how much data one session can a key can protect.

It does nothing against the traffic that's already captured. So it doesn't help you there. And then C protects the key, not the captured traffic. An HCM is generally a good practice and stops key theft from the server. But the adversary here never touches your server. They recorded it on the wire, and they're protecting the key at rest, does not basically protect the cipher against someone who's already has it, who already has access to it. The difference here is this when you're talking to someone like in this situation, they've already captured it, they've stored it, they have it. You messing with stuff isn't going to change any of that.

It's just there, they are looking specifically for how do I get this information? So you have to migrate your key establishment to a post-quantum or hybrid mechanism because you have that 25-year requirement and you know it's nation-state stuff coming after you. Okay, URS CISO. Next question. URS CISO of a large manufacturer. The board has asked you to prepare an organization for post-quantum migration. The environment spans several hundred applications accumulated over 15 years, including vendor products, internally built systems, and embedded controllers on the plant floor. What should you do first? A build a cryptographic inventory of algorithms, key lengths, locations, and data owners. C. Require vendors to attest to the post-quantum roadmap at contract review.

C. Begin replacing RSA ECC with MLKEM, MLDSA in external facing systems, or D assess applications for cryptographic agility so algorithms can be swapped out later. Okay, so there your board's asking you to post quantum migrations. They want to have the applications. You have a bunch of applications that have accumulated over 15 years, including vendor products, and you have all these different embedded controllers, which we talk about in the manufacturing space. What should you do first? Again, first, what should you do? And the answer is A, built a cryptographic inventory of algorithms, key links, and locations and data owners. Again, if you don't, you can't really protect what you don't know is there.

So that's the big factor. If you hit that rascal hard, right? So why are the other ones wrong? C, prioritizing before you have scope. Externally facing systems are visible, right? They're not necessarily the urgent one. You cannot prioritize them by data lifetime. C D, the right step but the wrong position. Assessing crypto agility genuinely matters, right? The systems cannot swap algorithms. They can't do that. That's the long pole. You need to figure those out, but you need to assess and understand why this is in factor. B, necessary and far too slow to be first. Contract renewal cycles run years, and this does nothing about the internally built systems or the embedded controllers.

So again, those are all factors you need to be aware of. All right, so let's get into the last question. So you are a defense contractor. A critical products cryptographic module holds a FIPS 140-2 certificate and went historically in September. So the vendor's FIPS 140-3 submission is in validation queue, which many are. Current validation timelines exceed 500 days. Your CMMC assessment window opens in four months. Your team has confirmed that no product holding an act no product holding an active FIPS 140-3 certificate meets the functional requirements. What is the most appropriate action? Okay, so you've got a defense contractor, you've got a dealing with 140-2, and you've got it, you're in the submission queue and it's going to take forever.

So what is the most appropriate action? A. Replace the product with one holding an active FIPS 140-3 certificate before assessment. B. Confirm the product runs FIPS mode and present that as evidence to the compliance people. C. Document the gap as a formal risk acceptance with executive sign-off and evidence remediation plan. Or D request the assessment to be postponed until the vendor's validation completes. Okay, so again, thinking like a manager, which one would you say it is? Yes, the answer is C. Document the gap as a formal risk acceptance with executive sign-off and evidence remediation plan. You want to make sure that this is all set up and make sure it's running with the uh and people understand the aspects related to this sign-off and that it's being covered.

This is an important part of all these organizations is that you have a good plan related to it. And just because they understand these situations, but if you have a plan, you have it documented, you are good. So why the others are wrong? The STEM already excluded it. The final sentence states that no valid alternatives meet the functional requirements. They already said that that it's it's basically already excluded that situation. Presenting configuration as validation, the question one error wearing the manager's suit. The FIPS mode is not a validation, and offering it as evidence of compliance misrepresents your posture. Okay. And then D, not a treatment, and it's not yours to grant.

Postponement of a scheduling request to a third party, you can't you can't have the rights to do that. And that doesn't make any sense. So you just have to document it and you have to then go forward with the documentation. So as we go through all this, it was never about cryptography. Assertion is not evidence. Again, a posture you set about yourself is not a certificate someone else issue. Bigger is not different. When the threat bypasses a defense entirely, scaling that defense changes nothing. Making more, you know, a little bit is good, more is even better. That does not work here. Find it before you fix it, right?

You need to understand what is going first. That's before you do anything else, you need to have an inventory in place and then read the constraints. ISC2 puts limiting fact on the last sentence, then offers the answer you would have given without it. So none of those are cryptographic knowledge, they are exam reasoning. This is again the whole way to think about a manager framework. That's the important part about all of this. Okay, so that's

Closing And How To Support 37:11

all I have. If you want to go check out CISSP Cyber Training, I've got my accelerator program, my pro program, and my Sprint cohort. The Sprint cohort will be going live here in actually I'm thinking of October, but it's gonna be going live in November. So we'll have the uh wait list is out there right now. We will be releasing some notes on when it's gonna be starting, and that will be starting in November, around the middle of November. So if you're interested, put your name on the wait list for that. The accelerator programs, these are all self-study and pro as well, but all of those are available.

You go check them out at CISSP Cyber Training and see what can be there and available to you. Okay, thank you so much for joining me today, and I'm excited to see where you guys take this. But again, we'll catch you all on the flip side. See ya. Thanks so much for joining me today on my podcast. If you like what you heard, please leave a review on iTunes as I would greatly appreciate your feedback. Also, check out my videos that are on YouTube and just head to my channel at CISSP Cyber Training, and you will find a plethora or a conocopia of content to help you pass the CISSP exam the first time.

Lastly, head to CISSP Cyber Training and sign up for 360 free CISSP questions to help you in your CISSP journey. Thanks again for listening.