Microsoft warns of China-backed threats to US infrastructure. AI image of Pentagon explosion dips market. And much more.

This certainly raised my eyebrows a bit. We know that this goes on all the time, but interesting to see it called out like this. They seem to be compromising Fortinet Fortiguard devices for initial access, and route their traffic through compromised residential and small business routers.

This kind of misinformation is going to be a thing more and more often.

The connectivity problems were due to a corrupt config file for a built-in security service, that auto-updates regardless of whether or not you allow automatic updates.

PyPi instigated a temporary suspension of all new users and package uploads over the weekend. It wasn't that they were flooded with more malicious packages than usual, just that they were short-staffed for a few days. Still, it shows just how much work goes into this.

Geo-based security detections are often used to flag suspicious logins, like when a user might log in from Brussels at 9am and somehow from Moscow at 11am. In the ever-evolving arms race between defenders and attackers though, Microsoft says that it's now more and more common for attackers to acquire residential IP addresses to prevent those alerts.

Patch 'em if you got 'em!

It apparently actively tries to guess the used password in order to scan the contents. Quite the balancing act between privacy and protecting customers.

Very nice. Before any Rust dependencies are used at Google they are thoroughly audited for security issues and supply chain risk. Those audits will now be shared with the outside world. HN thread here.

Interesting article that goes over the lack of clear disclosure rules for CISO's. The recent courtcase with the Uber CISO gave rise to a new set of worries around liability.

This is one for the bookmarks, should you ever face this situation. It nicely lists up which (US) agencies to contact and why. Related, CISA updated its ransomware guide this week as well.

Quite the story. His employer was under attack by a ransomware gang. As a Security Analyst he was on the incident response team, but he decided to switch the attacker's wallet address with his own and tried to convince his employer to pay.

Very interesting report on the current landscape of AI voice cloning and how it impacts cybersecurity. The link above goes to the summary, the report itself can be found here.

Remembering one strong password isn't all that difficult, but there is still the risk that it might be phished or keylogged. Passkeys on the other hand remove that risk entirely, and 1Password will soon allow you to use a passkey to unlock your vault. Very exciting stuff. (Sponsored)