decrypted · 16 november 2026 · vulnerabilities and patching · uk policy and law · ai and llm security
The Lords say the government is complacent about cascading crises
Peers have told the government it looks complacent about the risks that matter most to UK organisations, and they did it eight days before a vote on the Cyber Security and Resilience Bill, according to The Record. The House of Lords National Resilience Committee published its report on Monday. It is not a cyber report as such, but its argument is a Secure by Design one: Britain plans for single events, and attackers do not oblige.
What the committee found
The report, titled "Are we ready? Building a culture of resilience", follows a nine-month inquiry. Its central finding is that emergency plans are built around one crisis at a time, rather than several arriving together. The chair, Baroness Coussins, said that "the Government appears complacent, and the UK is underprepared."
The recommendations are structural. The committee wants a dedicated Cabinet minister for resilience, a chief resilience officer at permanent secretary level, and an annual prime ministerial statement to Parliament. It also wants a national resilience strategy within six months.
Think of a building with one fire door
Imagine an office with excellent fire doors, each tested separately. Now imagine a fire, a power cut and a blocked car park on the same morning. Every door works, yet nobody can leave. That is the committee's point about cascading failure. A supplier outage, a ransomware infection and a cable fault do not queue politely, and a business whose continuity plan covers each one alone has not designed for the day they coincide.
Secure by Design asks you to build the dependency map before the incident, not during it. Which three suppliers could stop you trading? What happens if two fail at once? Those are design questions, and they cost a few workshops rather than a budget line.
The bill and the insurance gap
According to The Record, the committee says the bill would set useful baseline requirements but its scope is limited, and it urges ministers to widen cyber regulation and fund regulators to enforce it. Many critical services sit in private hands: one outlet reports the evidence as showing more than 80% of UK critical national infrastructure is privately owned.
On insurance, the committee suggests the Flood Re model, the state-backed scheme that keeps home flood cover available, could be adapted to cover risks that are hard to insure, including large-scale cyber attacks. One outlet reports that recommendation so far, so treat the detail with care. The Cabinet Office said it would review the report and that there is "still more to do".
What this means for you
Nothing here is law yet, and ministers have only committed to a review. Opinion, not advice: write down your three most dangerous dependencies this week, test a scenario in which two fail together, and check whether your cyber insurance excludes the large, systemic events that worry the committee most. If it does, you have found a gap that no patch will close.
Also this week
AI-driven attack tooling. The FBI published an alert on 13 November about ARTEX, an open-source framework that uses a large language model (LLM), the kind of AI behind chatbots, to automate intrusions. The FBI says a variety of malicious actors have used it since at least August 2026, and that some intrusions went from first targeting to privileged access in under three hours. Its developer pulled the code from GitHub on 8 October, but forks, archived copies and a commercial edition keep it available, and researchers had found over 1,600 exposed ARTEX consoles by early November. The FBI says the tool is good at finding overlooked systems such as partner and supplier portals. The design lesson is dull and old: when attackers move at machine speed, the forgotten login page matters more than the flagship app. Know what you expose to the internet, and switch off what nobody owns.
A Linux flaw that crashes servers. Researcher Hyunwoo Kim disclosed NoisyTCP, tracked as CVE-2026-99288, an 11-year-old bug in the Linux kernel's TCP networking code, according to BleepingComputer. Red Hat describes a race condition that lets a remote attacker trigger a kernel crash, rates it Important and gives a preliminary CVSS score of 7.5. It is a way to knock a server over, not a way in, and Kim says the timing window is narrow, so a quiet server can take a long time to trip. BleepingComputer says a fix was merged upstream two weeks ago; check your distribution's advisory. Red Hat's workarounds are turning off a flood-protection setting, which weakens defences elsewhere, or limiting the network card to one receive queue. Patch the kernel instead where you can.
Sources
- The Record: British government 'complacent' on cyber threats from hostile states, peers warn
- Bytes Europe: Lords call for Flood Re-style safety net for cyber as they warn UK is unready for cascading crises
- FBI FLASH: Threat Overview of ARTEX Autonomous Penetration Testing Framework
- BleepingComputer: New NoisyTCP vulnerability lets attackers crash Linux servers
- Red Hat: CVE-2026-99288
If you want a second pair of eyes on your supplier dependencies, get in touch.
More like this
- Java4Shell: a decoding flaw that can let strangers run code on Java servers 18 november 2026
- The second flaw that turned a Check Point foothold into root 14 november 2026
- Barracuda's mail gateway has gone from patched to under attack 11 november 2026
Get the next post by email: subscribe to Decrypted. Double opt-in, unsubscribe any time, or take the RSS feed.
Prefer to listen? Decrypted on Apple Podcasts, or paste the podcast feed into any app.