Would you still use ChatGPT if your boss, or the AI provider, could read every single prompt you typed in? Most of us type something we would never share publicly into an AI tool every day.

In this Threat Talks Deep Dive, Field CTO Rob Maas sits down with ON2IT senior developer Derk Bell to unpack the privacy problem hiding inside everyday AI use, and a genuinely clever way to solve it.

The problem is bigger than a single prompt. An AI request carries far more context than a Google search: your conversation history, your files, and (with agents and MCP) automatic, autonomous access to your CRM, your email, your calendar and your codebase. A lot of that data is shipped off to a remote provider, often without you realizing it. Today we just trust those providers not to misuse it. As Derk points out, “trust us” is the exact opposite of Zero Trust: never trust, always verify.

So how do you verify? Derk walks through Confer, the privacy-first AI service from Signal Protocol co-creator Moxie Marlinspike. Using the analogy of a tamper-proof, locked calculator box for an “anonymous” employee survey, he explains how Confer lets you actually check the wiring: a hardware Trusted Execution Environment that isolates and encrypts the running program, remote attestation that lets you verify the exact open-source code handling your prompt, a Signal-style encrypted channel to talk to it, and hardware-bound keys that are thrown away the moment your session ends, so even the operator can never read your data.

We close on who needs this most (developers protecting trade secrets, executives working on M&A or strategy, and anyone bound by GDPR) and whether the big AI labs will adopt this the way the whole industry adopted Signal.

The takeaway: privacy and powerful AI do not have to be a trade-off. “Trust us” was never a security strategy.

🔗 Episode resources, transcript and show notes: https://threat-talks.com
📊 Grab the infographic referenced in the episode in the show notes to follow the technical steps.
🎙️ Subscribe to the podcast on Spotify, Apple Podcasts, or your podcast app of choice.

Threat Talks is a podcast by ON2IT cybersecurity and AMS-IX. We delve deep into the dynamic world of cybersecurity, one episode at a time. New episode every Tuesday.

Chapters:
00:00 Would you still use ChatGPT if your boss read every prompt?
01:00 Why AI is different from a Google search: context
02:08 "Trust us" and why that is the opposite of Zero Trust
03:22 It gets worse: AI agents, MCP and autonomous access
05:52 Enter Moxie Marlinspike, Signal, and Confer
06:30 The analogy: an "anonymous" employee survey in a locked box
10:16 Inside Confer: the TEE, attestation and the encrypted channel
14:21 End the session, throw away the keys
16:08 Who needs this? Developers, executives, GDPR-bound teams
18:29 Wrap-up: privacy and powerful AI can coexist

Ten years after its first release, the NIST Cybersecurity Framework got a full rebuild. Not because it failed, but because everyone started using it, and it was never built for everyone.

In this episode, host Lieuwe Jan Koning talks with Amy Mahn, IT Standards Advisor at NIST, and Daniel Eliot, Lead for Small Business Engagement at NIST’s Applied Cybersecurity Division, about what CSF 2.0 actually changes and why it took a multi-year public process to get there.

The original framework was written with critical infrastructure operators in mind. A decade later, hospitals, school districts, and small businesses were all using it too, often without the staff or budget the framework quietly assumed they had. NIST collected more than 4,000 comments from organizations across over 100 countries to find out what was missing.

The result includes a new Govern function that puts cybersecurity risk decisions in front of leadership and the board, a deliberately technology agnostic and vendor agnostic design that keeps the framework useful no matter which tools an organization runs, a Quick Start Guide aimed at organizations without a dedicated security team, and a sharper focus on supply chain risk.

Amy and Daniel also explain why “vendor agnostic” is a feature, not a gap: it’s what lets an organization decide for itself who or what is best suited to close a given risk, instead of a framework quietly picking winners.

Part 2, “CSF 2.0: Beyond the Framework,” continues the conversation and covers implementation in practice.

Threat Talks is a podcast by ON2IT cybersecurity and AMS-IX. New episode every Tuesday. Follow Threat Talks to stay up to date on the topic of cybersecurity.

Chapters:
00:00 Framework fatigue, the problem NIST heard
01:12 Welcome to Threat Talks
02:05 Meet Amy Mahn and Daniel Eliot (NIST)
03:40 What the original CSF got right, and where it broke down
07:15 Four thousand comments, one framework
10:30 The new Govern function
14:05 Why CSF 2.0 is technology agnostic and vendor agnostic
17:20 The Quick Start Guide for teams without a security team
20:10 Supply chain risk gets its own seat at the table
22:45 What Part 2 will cover
24:48 Closing thoughts

Agentic AI is powerful, and someone recently found that out the hard way when an AI tool, given free rein with a user’s own permissions, deleted her entire mailbox. That cautionary tale opens this Threat Talks Deep Dive, where host Lieuwe Jan Koning talks with Rob Maas, Field CTO of ON2IT, about what Zero Trust looks like when the thing you’re securing is an AI agent.

Drawing on Rob’s recent blog post (and the Zero Trust pillars shared by CISA and Forrester’s Zero Trust eXtended framework), they work through each pillar in turn. The recurring theme is “just-in-case” privileges: the broad access we hand humans on the assumption they’ll use judgment. Agents have no such judgment. Give one an intent and it will use everything it has to reach the goal, and it can spin up parallel instances to get there faster.

Across Identity, Devices, Network, Applications & Workloads, and Data, Rob makes the case for:

• Non-human identities with just-in-time, quickly-rotated privileges, so a leaked token can’t be reused forever.
• Tightly constrained execution environments (VM, container, serverless) that only touch what the agent truly needs.
• Identity-based network segmentation, so an agent working with CRM data can never reach the financial system.
• Allow-listed MCP tooling, because tool sprawl is the new shadow IT.
• New data controls for a world where everything (prompts, retrieval, documents) is data flowing to and from a model.

He’s candid about the gaps, too: there’s no generic “AI firewall” yet, prompt injection has no guaranteed fix, and the hardest control points now live in the details of how individual developers configure their tools. The optimistic note: because agent-to-model and agent-to-agent calls can be logged, you can actually see what an agent is doing, an advantage over the opacity of the human mind. The episode closes on what’s still missing and a clear first step for any organization: get an overview of every agent and MCP server in use, and the access each one has.

Threat Talks is a podcast by ON2IT cybersecurity and AMS-IX. New episode every Tuesday. Follow Threat Talks to stay up to date on the topic of cybersecurity.

Mythos found a 23-year-old vulnerability in FreeBSD that no human team had caught. Your 30-day patch cycle assumes years before it gets weaponized. Today that window is one day. Next year it will be one hour.

Lieuwe Jan Koning, Co-founder & CTO at ON2IT, sits down with Rob Maas, Field CTO at ON2IT, to break down what Anthropic's Mythos actually found, why the public release (Fable) still frustrates security professionals, and whether the FABLE framework gives defenders a realistic path forward.

Rob's verdict: there is truth in what Anthropic claims. It is not as catastrophic as the marketing suggests. But if your fundamentals are not in place, the time to fix that is now.

00:00:00 Introduction
00:00:46 What is Mythos? From Project Glasswing to Fable
00:03:13 What Mythos actually found: FreeBSD, Palo Alto, real patches
00:05:57 The zero-day clock: from years to one hour
00:09:00 The FABLE framework and the CSA "Mythos Ready" paper
00:15:24 Authentication, segmentation, and egress filtering
00:20:51 Myth or reality: Rob's verdict

Subscribe to Threat Talks and turn on notifications for deep dives into the world's most active cyber threats and hands-on exploitation techniques.

🔔 Follow and Support our channel! 🔔
===
► YOUTUBE: / @threattalks
► SPOTIFY: https://open.spotify.com/show/1SXUyUE...
► APPLE: https://podcasts.apple.com/us/podcast...

👕 Receive your Threat Talks T-shirt
https://threat-talks.com/

🗺️ Explore the Hack's Route in Detail 🗺️
https://threat-talks.com

🕵️ Threat Talks is a collaboration between @ON2IT and @AMS-IX

Every big nation state has a cyber army: China, Russia, the US, Europe. But what about Iran? Meet Boggy Serpens, a group tied to Iran’s civilian intelligence service whose entire business is breaking in and staying in, then handing the keys to whoever strikes next.
Their playbook, Operation OLALAMPO, needs just one booby-trapped Word document to plant three separate backdoors on your network.

• A Telegram-bot command channel that hides inside everyday encrypted chat traffic, a Rust “Ghost” backdoor built to defeat analysis, and a legitimate AnyDesk install quietly turned against you.
• The layered defense for every stage: email and file controls, behavioral EDR, egress policy, threat intel, and Zero Trust segmentation.
• The twist: why this operation mostly failed, plus the tells that the malware was partly written with AI.

Filmed live at the ON2IT SOC, host Lieuwe Jan Koning runs a red team vs blue team session with analysts Yuri Wit, the “proxy Iranian” attacker, and Rob Maas on defense. Watch the full episode to see each move, and the exact control that stops it.

🔗 Episode resources, transcript and show notes: https://threat-talks.com 🎙️ Subscribe to the podcast on Spotify, Apple Podcasts, or your podcast app of choice.

Threat Talks is a podcast by ON2IT cybersecurity and AMS-IX. We delve deep into the dynamic world of cybersecurity, one episode at a time. New episode every Tuesday.

Chapters (paste into YouTube description)

00:00 Every nation state has a cyber army: what about Iran?
00:21 Meet the guests: Yuri (red team) and Rob (blue team)
01:17 Boggy Serpens and Operation OLALAMPO: Iran's access brokers
04:20 Infection via Office macros, and the social-engineering layer
07:18 You opened the document: three payloads
08:06 Backdoor 1: Telegram-bot command and control
12:55 Backdoor 2: the Rust "Ghost" backdoor, and why it's so hard to analyze
16:03 Backdoor 3: legitimate AnyDesk, pre-loaded for the attacker
17:59 Zero Trust and network segmentation
18:59 Did it work? AI tells, and staying vigilant

In 2026, 40 new submarine cables go live. Most won't land in Europe. Europe is losing the sea cable race, and most people haven't noticed yet.

In this second part of our sea cables conversation, host Peter Ernst sits down with Ernst Noorman, the Netherlands' Cyber Ambassador-at-Large and a member of the ITU Advisory Body on Submarine Cable Resilience, to move from the “how” of sea cables to the “why it matters.”

We compare two places that were once called the two hardest spots in the world to build digital infrastructure, Amsterdam and Singapore, and unpack how Singapore solved its crunch with 32 cable landings, five years of zero cable faults, and a green-energy-first tender process, while the Netherlands risks resting on a 30-year-old head start.

Along the way: the difference between sovereignty and autonomy, why “always the cheapest option” no longer works, the EU Cyber Resilience Act and security by design, what NIS2 means for boards and CEOs personally, and why Europe needs to stop being modest about Airbus-sized wins.

Chapters

00:00 — 40 new cables, most skip Europe
00:30 — Meet Ernst Noorman & the ITU advisory body
02:00 — The sea cable map is being redrawn
04:08 — Why the Netherlands risks losing its head start
06:26 — How Singapore solved it: 32 landings, zero faults
08:09 — Tax cuts for digital, would Europe ever?
08:59 — Sovereignty vs autonomy: it's about choice
15:02 — You can't own the whole stack (ASML, Nokia, Ericsson)
15:53 — Why “always the cheapest” stops working
17:47 — The Cyber Resilience Act & security by design
18:51 — The water-from-the-tap analogy
19:51 — What boards and CEOs must actually ask
25:30 — Back to Singapore: government-led, by design
29:39 — The good news: Europe's real strengths
36:15 — What needs to happen in the next 3–5 years

Threat Talks is a podcast by ON2IT and AMS-IX. Subscribe for more on Zero Trust, cyber resilience, and the infrastructure behind the internet.

The headlines say Russia’s shadow fleet is cutting cables. The experts say most faults come from clumsy ship anchors. Ninety-nine percent of global internet traffic runs across the ocean floor, and the conversation about what threatens it is mostly wrong.

In this episode of Threat Talks, Peter van Burgel, CEO of AMS-IX, sits down with Ernst Noorman, Cyber Ambassador at Large for the Netherlands and member of the ITU Advisory Board on Submarine Cable Resilience, to separate geopolitical noise from engineering reality, and explain what actually puts global internet connectivity at risk.

Timestamps

00:00:00 Introduction
00:00:55 The ITU Advisory Board on Submarine Cable Resilience
00:05:04 Shadow Fleets, Geopolitics, and the Sabotage Myth
00:10:30 Shunts, Faults, and What Actually Breaks Cables
00:15:47 Why Satellite Cannot Replace Submarine Cables
00:17:06 Digital Sovereignty and the Big Tech Cable Takeover
00:28:16 What Every CEO Should Put on the Agenda

Key Topics Covered

• Why most submarine cable faults come from anchors, fishing nets, and natural events, not state actors

• How aging repair ships and bureaucratic permitting barriers make restoration slow in most of the world

• Why satellite (including Starlink) cannot replace subsea fiber at any meaningful scale

• How big tech dominance over new cable investment creates digital sovereignty risks for governments and large organizations

• What NIS2 means for CEO accountability on digital infrastructure resilience

Related ON2IT Content & Referenced Resources

ITU Advisory Board on Submarine Cable Resilience: https://www.itu.int/digital-resilience/submarine-cables/advisory-body/
ICPC (International Cable Protection Committee): https://www.iscpc.org
Dutch Cybersecurity Council / CEO manual on NIS2: https://www.cybersecuritycouncil.nl
Dutch Cybersecurity Act (NIS2 implementation): https://www.dutchncca.nl/the-cybersecurity-act

Threat Talks: https://threat-talks.com/russia-cutting-cables-whos-protecting-it/
ON2IT (Zero Trust as a Service): https://on2it.net
AMS-IX: https://www.ams-ix.net/ams

Subscribe to Threat Talks and turn on notifications for deep dives into the world’s most active cyber threats and hands-on exploitation techniques.

🔔 Follow and Support our channel! 🔔

► YOUTUBE: / @threattalks
► SPOTIFY: https://open.spotify.com/show/1SXUyUE…
► APPLE: https://podcasts.apple.com/us/podcast…

👕 Receive your Threat Talks T-shirt https://threat-talks.com/

🗺️ Explore the Hack’s Route in Detail 🗺️ https://threat-talks.com

🕵️ Threat Talks is a collaboration between @ON2IT and @AMS-IX

A company skips a security check two days before Black Friday and loses $1 million when transactions land in the wrong bank accounts. A machine learning team is told no on production data access, gets it via SharePoint anyway, and a year later the data is on contractor laptops nobody can account for.

Two stories, one pattern: when security blocks, the risky work doesn’t stop – it just happens without you.

Lieuwe Jan Koning, Co-founder and CTO at ON2IT Cybersecurity, sits down with Sina Yazdanmehr, Founder and Managing Director of Aplite GmbH, on the prevention paradox, why a “no” from the CISO is an illusion of control, and how a technical security team turns into a business partner instead of a roadblock.

Timestamps

00:00:00 Introduction
00:01:55 The $1 million Black Friday story
00:04:14 Hero culture rewards shipping, not prevention
00:06:55 The prevention paradox
00:08:00 NIS2 and executive accountability
00:09:00 Avoiding the Department of No
00:12:18 Production data on contractor laptops
00:16:13 The technical CISO as business partner

Key Topics Covered

• Why hero culture quietly trains organizations to bypass security under deadline pressure
• The prevention paradox: why the person who avoids a loss never gets the credit
• What happens after a CISO says no: shadow workflows, friendly handovers, and data on laptops nobody owns
• What a counter-proposal in risk-based language gets you that a flat refusal does not

Related ON2IT Content & Referenced Resources

• Aplite GmbH: https://aplite.de
• Previous Threat Talks with Sina Yazdanmehr (Security Culture part 1): https://youtu.be/1JnAsXDCKzM?si=qFlMxC617E30U1dW
• Previous Threat Talks with Sina Yazdanmehr: https://www.youtube.com/watch?v=wBodTl_nY1w
• Previous Threat Talks with Sina Yazdanmehr: https://www.youtube.com/watch?v=fBwdGXf-0dY
• Threat Talks: https://threat-talks.com/
• ON2IT (Zero Trust as a Service): https://on2it.net/
• AMS-IX: https://www.ams-ix.net/ams

Subscribe to Threat Talks and turn on notifications for deep dives into the world's most active cyber threats and hands-on exploitation techniques.

🔔 Follow and Support our channel! 🔔
===
► YOUTUBE: / @threattalks
► SPOTIFY: https://open.spotify.com/show/1SXUyUE...
► APPLE: https://podcasts.apple.com/us/podcast...

👕 Receive your Threat Talks T-shirt
https://threat-talks.com/

🗺️ Explore the Hack's Route in Detail 🗺️
https://threat-talks.com

🕵️ Threat Talks is a collaboration between @ON2IT and @AMS-IX