Enterprises believe they have container security under control—until they discover their base images ship with 300 vulnerabilities and 70 high-critical CVEs before a single line of application code is written. The problem isn’t lack of visibility. It’s that DevSecOps teams are starting from a fundamentally broken foundation, spending thousands of engineer-hours patching commoditized infrastructure that should have been secure by default.

The Guest: John Morello, CTO and Co-Founder at Minimus

Key Takeaways

• Pull the latest official Python container image today and you start with 300 vulnerabilities—70 high and critical—before writing application code
• Software Bill of Materials (SBOM) documents what vulnerabilities exist but doesn’t reduce risk; Cryptographic Bill of Materials (C-BOM) declares post-quantum crypto readiness
• Adversaries are executing “harvest now, decrypt later” attacks—recording encrypted data today for future decryption when quantum computers become viable
• Pre-built OCI-compliant container images from upstream source eliminate 97% of vulnerabilities and reduce engineering overhead on non-differentiated work
• NIST SP 800-190 provides the gold standard framework for container security from hypervisor to application layer

***

[expander_maker]

In a recent TFiR interview at KubeCon + CloudNativeCon Amsterdam, Swapnil Bhartiya spoke with John Morello, CTO and Co-Founder at Minimus, about the evolving container security landscape, the urgent post-quantum cryptography transition, and why starting with secure-by-default infrastructure is the only scalable path forward.

The Container Security Paradox: Vulnerability Overload at the Foundation

Morello, who pioneered container security at Twistlock (acquired by Palo Alto Networks for Prisma Cloud), identified a fundamental problem in how organizations approach container security: they’re starting from a broken foundation and spending engineering resources on commoditized, non-differentiated work.

Q: How has KubeCon’s focus on security evolved since the early days?

John Morello: “Our first company, Twistlock, started the whole container security space. We were at the very first KubeCon. With Twistlock, what we tried to do was primarily help customers understand what vulnerabilities they had and protect their containers at runtime. But with Minimus, what we’re trying to do is solve that problem before it ever happens in the first place.”

Minimus provides a library of hundreds of pre-built, OCI-compliant container images built continuously from upstream source, allowing organizations to replace their existing base images and reduce vulnerabilities by over 97% simply by changing the initial image reference.

Q: Why can’t organizations just maintain these images themselves?

John Morello: “Even if your team can do that, you’re spending countless man-hours on something that’s a completely commoditized, non-differentiated task. It doesn’t improve your customers’ or your constituents’ experiences. If you can offload that work to us, we’ve built this whole pipeline to do that effectively at scale, and you can get the same end result much more efficiently.”

He emphasized that almost no organization does this well today unless they’re a Meta or Netflix with access to top-end engineering talent focused exclusively on platform infrastructure.

Q: What’s the starting point vulnerability count for standard base images?

John Morello: “If you were to pull the very latest Python image today—the very latest official version of Python—you’re going to get something with almost 300 vulnerabilities, including 70 that are high or critical. You’re never going to be able to have a good outcome if your foundation starts off like that.”

Why SBOM Alone Doesn’t Reduce Risk

Morello distinguished between visibility and risk reduction, explaining that Software Bill of Materials (SBOM) is a necessary but insufficient component of supply chain security.

Q: Is having an SBOM enough for supply chain security?

John Morello: “SBOM is a great capability to provide a consistent way to describe what is inside a container image. But it’s not sufficient to help you actually reduce the risk. Knowing that you have vulnerable components doesn’t remove the risk, it just tells you that you’ve got the problem.”

Minimus addresses this by providing both vulnerability-free images and cryptographically verifiable SBOMs that allow organizations to prove deterministically using open source tools like Sigstore and Cosign that specific components and versions are present and vulnerability-free.

Cryptographic Bill of Materials: Preparing for the Post-Quantum Transition

Morello detailed the emerging Cryptographic Bill of Materials (C-BOM) standard and why NIST is integrating it into the SBOM specification to address the post-quantum cryptography transition.

Q: What is a C-BOM and why does it matter?

John Morello: “A C-BOM stands for cryptographic bill of materials. Conceptually, it’s very similar to an SBOM. There was a realization in the industry that it’s very hard to deterministically know what cryptographic algorithms exist in something if you don’t have a declarative way for the creator of that artifact to tell you what they are. A C-BOM would allow you to see inside an image the cryptographic frameworks, key lengths, and algorithms available.”

He explained that later in 2026, the SBOM standard is expected to incorporate what had been a separate C-BOM standard, making it easier for organizations to understand their cryptographic posture and track progress toward post-quantum readiness.

Q: What does NIST’s post-quantum cryptography project mean for container security?

John Morello: “NIST has been working on post-quantum cryptography for a number of years now. Changing cryptographic algorithms is one of the most challenging software updates you can do in the ecosystem because it has implementations across Linux, Windows, web servers, databases, and often extends all the way down to the silicon, with ASICs designed to accelerate cryptographic operations.”

Minimus already offers FIPS 140-3 validated images with CMVP certificates and plans to support the forthcoming FIPS 140 variant for post-quantum compliance as NIST finalizes those standards.

The “Harvest Now, Decrypt Later” Quantum Threat

Morello emphasized the urgency of the post-quantum cryptography transition by explaining the adversarial strategy of recording encrypted data today for future decryption.

Q: Why is post-quantum cryptography urgent if viable quantum computers don’t exist yet?

John Morello: “If someone is recording data today that’s using non-quantum-resistant algorithms, whoever has recorded that data could theoretically decrypt it in the future. That’s why you can’t wait until there’s a viable quantum computer before you start dealing with this issue, because it takes years to develop the algorithms. If your data today is going to be relevant 10 years from now, it’s already compromised because those actors have already gathered the data and will harvest it.”

He compared the quantum threat to the rapid disruption caused by ChatGPT and OpenAI in the AI space, noting that breakthrough advances in quantum computing could arrive suddenly and without warning.

Q: What’s the first step organizations should take toward post-quantum readiness?

John Morello: “The first step in almost any case like this is just knowing where you are today. C-BOM is a relatively new standard. You probably have lots of stuff that has no C-BOMs published for it. Try to get an idea across your environment—what versions of Red Hat, Debian, SSL accelerators, or silicon you have—and then figure out with your vendors what their plans are for releasing post-quantum algorithms.”

NIST SP 800-190 and the Layered Container Security Model

Morello referenced NIST Special Publication 800-190 as the gold standard for container security best practices, describing how C-BOM and SBOM fit into that framework.

Q: How does C-BOM fit into NIST’s container security framework?

John Morello: “NIST has historically provided best practices and frameworks for securing different parts of environments and technology stacks. For us, 800-190 describes the best way to think about container security in a layered approach from the hypervisor all the way into the application and across the lifecycle. C-BOM fits into that because it’s the way you can show that you’ve configured that image to align with those best practices.”

Without a standardized way to describe cryptographic configurations, he noted, it’s extremely difficult to know across an enterprise whether security is being implemented consistently and where outliers exist.

Starting Secure by Default: The Minimus Approach

Morello articulated the philosophy behind Minimus: reduce both risk and operational overhead by providing secure-by-default infrastructure that follows industry best practices from the start.

Q: What makes starting with a secure foundation so critical?

John Morello: “If you’re the CISO, you’re probably dealing today with enormous amounts of alerts about vulnerabilities that your team is not directly responsible for. If you can start off with a better foundation, it’s much less work, much less risk, and much less hassle for your colleagues who have to deploy and update this software if they just avoid those problems to start with.”

He emphasized that Minimus images don’t just eliminate CVEs—they also include the most modern cryptography available, follow NIST and CIS benchmarks for container and application configuration, and are hardened by default so that security teams don’t have to think about how to remove legacy crypto or harden images post-deployment.

Q: What tooling should organizations use to trace vulnerabilities from production back to source?

John Morello: “Having good tooling that spans from the beginning of your development lifecycle all the way through production—products like Wiz, Orca, and Prisma Cloud—allows you to see if you have a vulnerability in production and trace it all the way back to the Git repository where that vulnerability was originally created. If you fix those things upstream instead of fixing 1,000 instances downstream, you can immediately resolve all those issues downstream.”

That upstream-first philosophy is fundamental to how Minimus operates: fix vulnerabilities at the foundation and eliminate thousands of downstream alerts automatically.

The Urgency of Acting Before Quantum Disruption

Morello concluded by reinforcing the urgency of preparing for post-quantum threats, drawing a parallel to the sudden AI disruption that reshaped the industry in a matter of months.

Q: How should organizations think about the timeline for quantum threats?

John Morello: “There’s so much research going into quantum computing that it seems fairly likely that there will eventually be viable quantum computers capable of effectively and almost instantaneously breaking all the current forms of cryptography we have. The reason you’re encrypting data is probably because you’re trying to pass it over untrusted media like the internet. If someone can record all that data, even if today they can’t do anything with it, they can keep saving it. Eventually, when that quantum computer exists, they can decrypt all that data. That’s a real risk, and that’s why it’s not something you can wait for until it happens.”

Watch the full TFiR interview with John Morello here.

[/expander_maker]