Nevada’s ransomware saga unfolded like a tech thriller, with a trojanized tool, a persistent hacker, and a 28-day recovery marathon. Instead of paying ransom, the state shelled out $259,000 in overtime to restore systems. The report highlights transparency in cybersecurity incidents—something as rare as a unicorn!

Google sounds the alarm on self-modifying AI malware. This new breed of malware uses AI to mutate and adapt in real-time, making it harder to detect. Dubbed “AI-in-the-loop,” it evolves during execution, marking a significant shift in cyber threats that even seasoned defenders might find challenging to tackle.

Cisco warned about a new attack variant targeting its firewalls, causing them to reload and leading to denial-of-service conditions. Despite patches, these firewalls have been under attack for months. Meanwhile, two critical bugs in Cisco Unified CCX software could allow attackers to execute commands with root privileges. Time to patch up, folks!

The EFF and AV Comparatives find stalkerware detection still a mixed bag. In 2025, Malwarebytes hit the bullseye, detecting 100% of stalkerware products. Meanwhile, Google Play Protect seemed to take a nap, catching just 53%. Despite fewer stalkerware products, it’s still a cat-and-mouse game for anti-virus apps.

Prompt injection in ChatGPT is like slipping a “kick me” sign onto the AI’s back without anyone noticing. Attackers can hide malicious instructions in blog comments or indexed websites, tricking the AI into following orders it shouldn’t. It’s a digital prank with serious consequences, highlighting ongoing AI security challenges.

Watch out, ChatGPT users! Tenable researchers have uncovered seven new vulnerabilities in ChatGPT that could lead to data theft. From sneaky “bio” features to devious “open_url” commands, these attack techniques are no joke. Even GPT-5 isn’t entirely safe from these crafty prompt injections, proving AI still has some learning to do.

Cybercriminals infiltrated Booking.com partner accounts, unleashing chaos with a sophisticated malware campaign. Using social engineering, they tricked hotel staff into executing commands that installed the PureRAT remote access Trojan. Victims were duped into paying twice for their hotel stays. The campaign highlights the increasing professionalization of cybercrime targeting the travel industry.

InedibleOchotense is impersonating ESET in phishing attacks against Ukrainian entities, using trojanized installers to deliver malware. Meanwhile, Sandworm continues to wreak havoc with wiper attacks, and RomCom exploits a WinRAR vulnerability in spear-phishing campaigns, proving that when it comes to cybersecurity, Russia-aligned threat actors are anything but “remotely” funny.

Cisco urges users to patch Cisco Secure Firewall software immediately to avoid unexpected reloads and potential DoS conditions. New attack variants exploiting CVE-2025-20333 and CVE-2025-20362 can wreak havoc, so update now! Also, remember to address those pesky Unified CCX flaws to keep your network safer than a paranoid squirrel’s nut stash.

Open-weight large language models may ace single-turn defenses, but throw in a multi-turn adversarial attack, and they’re as defenseless as a chocolate teapot. Cisco AI Defense found that persistent conversations can outwit robust defenses, proving that models need more than a stiff upper lip to handle iterative manipulation.

In the latest plot twist, ClickFix attacks now have video tutorials to guide you through your own cyber demise, complete with a countdown timer to ensure you panic just enough. Push Security highlights these scams that auto-detect your OS for a personalized malware experience. Remember, no legit verification process requires terminal code execution.

Truffle Security, known for its secret scanning tool TruffleHog, has secured $25 million in Series B funding. With backing from Intel Capital and a16z, the San Francisco-based company plans to enhance TruffleHog Enterprise and launch a Google Cloud-focused add-on. It’s a great day for secrets, unless you’re one of them!

Passwords like 123456 and admin are still reigning champions despite years of IT advice. A new study by Comparitech highlights our collective laziness with top offenders like qwerty123. Maybe it’s time to ditch these predictable choices. Pro tip: try “icantbelivewerestilltellingy0uthis” to keep hackers guessing. Stay safe and avoid the common passwords list!

Cisco has released crucial security updates for its Unified Contact Center Express (UCCX) software, humorously dubbed a “contact center in a box.” The flaw, CVE-2025-20354, lets attackers execute commands with root privileges. So, unless you want hackers playing puppet master with your systems, upgrading is the punchline here.

Agentic AI shifts the conversation from AI as passive tools to autonomous systems that plan, act, and coordinate. This evolution promises operational speed but also introduces new risks. Autonomy can lead to unintended actions, creating governance challenges that business leaders can no longer ignore.

Cybercrime is no longer confined to the virtual realm; it’s creeping into real life. From organized crime funding to renting violence as a service, the digital world is spilling its chaos into our everyday lives. Stay informed, keep learning, and remember: cybersecurity keeps changing fast, so our understanding must keep up.

SonicWall’s recent breach involved a state-sponsored group accessing firewall backup files via an API call. SonicWall faces geopolitical cyber threats but insists it’s now “stronger, more resilient, and even more trusted.” The company emphasizes the breach didn’t compromise their firewall software, just a supporting cloud service.

Hackers drained over $120 million from Balancer by exploiting a rounding function and batch swaps, proving once again that math mischief can be quite profitable. While Balancer scrambles to recover funds, they might want to consider swapping their rounding function for one that doesn’t round down to “broke”.

ENISA reports that hacktivist-driven DDoS attacks accounted for 60% of public sector cybersecurity incidents last year. While DDoS attacks were frequent, data threats proved more disruptive. With public administrations now in the “risk zone” for compliance, ENISA urges improved resilience against these cyber threats to protect vital services and sensitive data.

Scam ads are taking over social media, but two former Meta staffers are on a mission to save the day. They’ve launched a nonprofit to shine a light on this murky world, aiming to bring transparency and help users dodge digital deception. Who knew fighting scams could be this enlightening?