Anthropic says Mythos has found thousands of critical bugs, hackers leak documents from a Russian disinfo group, GitHub rolls out new npm security features, and Dutch police raid two bulletproof hosting providers.

In this sponsored interview James Wilson chats with Sondera CEO Josh Devon about why guardrails and instruction files aren’t enough to keep AI agents from going haywire. EDR, DLP and other traditional controls can’t and won’t prevent agents from going rogue.

Josh explains Sondera’s “principle of least autonomy” for agents: let them do useful work, but put them in a deterministic policy harness so they can’t leak secrets, abuse tools or wander off-task.

Microsoft ends support for SMS MFA on personal accounts, GitHub was hacked via a malicious VS Code extension, CISA will let researchers submit new KEV entries, and an SMS blaster was detained at Eurovision.

Tom Uren and James Wilson talk about moves from several European governments to ditch Signal and set up their own encrypted messaging systems for internal government use. These efforts are motivated by concerns about phishing and sovereignty, but the solutions being adopted are imperfect and will come with their own set of problems. Signal fills a space that can’t be filled with sovereign capability.

They also talk about Fast16 malware. We are only now learning about the second arm of a mid-2000s campaign to delay Iran’s nuclear weapons program that included the infamous Stuxnet worm.

This episode is also available on YouTube

Microsoft disrupts a malware-signing service used by ransomware gangs, a CISA contractor leaks sensitive GovCloud keys, vulnerability exploitation is now the dominant network entry vector, and Drupal readies security updates for a “highly critical” vulnerability.

In this edition of Between Two Nerds Tom Uren and The Grugq look at Department 4 of Bauman Moscow State Technical University where students learn how to hack for the state. Its curriculum is extremely explicit about how the hacking and propaganda operations are relevant to state operations. They discuss whether this is an advantage for Russia’s cyber program and look at what Western intelligence agencies do instead.

This episode is also available on YouTube.

Indonesia emerges as a new cyber scam hub, Grafana got hacked and held for ransom, the Fast16 malware subverted software used to simulate nuclear explosions, and a new Microsoft Exchange zero-day is under attack.

In this sponsored interview James Wilson chats with Push Security’s Chief Research Officer Jacques Louw about how the company has integrated an army of AI agents into its threat detection platform.

Not only has agentic AI led to the discovery of Install Fix campaigns, but it will help simplify the platform for new customers.

The source code for the Shai-Hulud worm has been released online, a dark web market admin was charged after a major OPSEC failure, France investigates an Israeli disinfo firm, and ‘Composer’ rushes to fix a GitHub token leak.

Tom Uren and James Wilson talk about the argy bargy within the Trump administration about AI regulation. They cover who is fighting, what is at stake and what the real areas of concern are.

They also cover low earth orbit satellite constellations. Russia’s building one, the EU has plans and China is building two. They are the new must-have accessory for any country with global ambitions.

This episode is also available on YouTube