Be careful what you post online. Here, "online" does not mean social media; it means online tools that you are likely to use in your day-to-day activities. It's also worth noting that cybercrime is a real threat and with the popularity of LLMs, cyber criminals can now automate lots of things, e.g., creation of phishing emails.
The security of your organization also depends on your cybersecurity education and awareness. This can potentially save your organization from financial losses or loss of personal information due to your unpreparedness for what can go wrong if you don't take certain steps to secure your digital devices.
Then we have attackers abusing Cascading Style Sheets (CSS) to track users! Yes. You read that right. If you ever thought CSS was for web development alone, think again.
Thousands of Secrets Leaked on Code Formatting Platforms
Please, before you use these formatting platforms, ensure that you're not leaking sensitive information. It does not hurt to double-check.
Now, who owns the leaked secrets? From the article:
The leaked secrets belong to organizations across multiple verticals, including technology and cybersecurity, critical national infrastructure, government, finance, healthcare, aerospace, insurance, banking, education, telecoms, travel, and more.
Account Takeover Fraud Caused $262 Million in Losses in 2025: FBI
To begin with, that's a lot of money. Also, it appears that the cybercriminals use social engineering in the attacks.
From the article:
As part of these attacks, the FBI notes, the miscreants convince the victim to share their login information, including multi-factor authentication (MFA) code or one-time passcode (OTP), to access their account.
Next, they log into the victim’s account on the financial institution’s website and reset the password to gain full control of the account and lock the victim out.
WormGPT 4 and KawaiiGPT: New Dark LLMs Boost Cybercrime Automation
While we have legitimate AI applications powered by LLMs, we have malicious ones as well. Examples are WormGPT 4 and KawaiiGPT. Besides being malicious, these LLMs reduce the barrier to entry for less skilled cyber criminals to almost nothing.
From the article:
While WormGPT 4 is advertised to users as a “key to an AI without boundaries”, Palo Alto researchers noted, “The developers of WormGPT 4 maintain secrecy regarding its model architecture and training data. The second dark LLM analyzed by Palo Alto researchers is KawaiiGPT, which appears to have emerged in July 2025. KawaiiGPT is freely available on GitHub and easy to set up.
Invisible tracking in the browser: how cascading style sheets (CSS) are abused
If I remember correctly, I read something similar on Bleeping Computer some years ago. Now, I was surprised when I read this article to learn that CSS can be used for more than designing web pages.
A quick one from the article:
Email clients render content differently and support different CSS rules and properties. At the same time, CSS can be used by attackers to track user actions and preferences. It is also possible to read the hardware and system “fingerprints” of recipients, but many of these approaches work only in specific email clients and depend on specific configuration assumptions.
CISA Warns of Active Spyware Campaigns Hijacking High-Value Signal and WhatsApp Users
Now, before you conclude that you are not a high-value target and you have no need to read the article, let me tell you something: read the best practices outlined at the end of the article.
For a start, here are some:
- Only use end-to-end encrypted (E2EE) communications
- Enable Fast Identity Online (FIDO) phishing-resistant authentication
- Move away from Short Message Service (SMS)-based multi-factor authentication (MFA)
Credits
Cover photo by Debby Hudson on Unsplash.
That's it for this week, and I'll see you next time.
Top comments (0)