Dragon Routers and AI Phishing: Why Beijing Wants Your Boring Water Utility More Than Your Secrets

Dragon's Shopping Spree: Beijing's 29-Minute Smash and Grab Hits AI Labs and Logistics Giants

This is your Digital Dragon Watch: Weekly China Cyber Alert podcast. Hey listeners, Ting here with your Digital Dragon Watch, and the last week in China cyber has been…busy. Let’s start with the big strategic picture. CrowdStrike’s latest reporting says China‑nexus hacking groups ramped intrusion activity by 38 percent in 2025, with an 85 percent spike against logistics companies and major pressure on tech and telecom.[2][5] Adam Meyers at CrowdStrike even called logistics “probably the top target” for Chinese threat actors. That trend hasn’t slowed this week: shipping, cloud providers, and undersea‑cable–adjacent networks are still getting hammered as Beijing tries to map and potentially disrupt global supply chains. The favorite new‑ish attack vector? Edge devices. According to CrowdStrike, roughly 40 percent of China‑linked exploits last year hit internet‑facing VPNs, firewalls, and gateways, and 67 percent of those bugs gave immediate system access.[2] Over the past few days, several US and European incident‑response teams have quietly flagged fresh compromises in unpatched VPN appliances at mid‑size cloud and telecom providers, tracking back to familiar China‑nexus clusters like Warp Panda and Phantom Panda mentioned in the CrowdStrike report. While those edge hits are quietly exfiltrating data, another thrust is pure AI theft. CrowdStrike’s “China Stealing the AI Tech It Can’t Build” analysis describes how Chinese operators are using cyberespionage as industrial policy to close the AI innovation gap.[5] In the last week, multiple US AI startups have reported targeted phishing and OAuth abuse against their MLOps platforms, mirroring techniques in that report: credential‑stuffing against admin dashboards, followed by rapid grab‑and‑go of model weights and training data. Breakout time is now averaging 29 minutes from first foothold to lateral movement.[2] That’s not hacking; that’s smash‑and‑grab with a stopwatch. On the defensive side, US government response is getting sharper. The FBI, through ongoing campaigns like Operation Riptide highlighted by FBI field offices, keeps reminding companies that state‑sponsored and criminal activity are blurring, and is leaning hard on rapid reporting of China‑linked intrusions.[10] CISA has been pushing joint advisories urging immediate patching of edge devices within 72 hours of disclosure, tighter network segmentation, and continuous monitoring for anomalous traffic from VPNs and firewalls—exactly the weaknesses Chinese actors are exploiting, according to CrowdStrike’s data.[2] So what should you do this week, not someday? First, treat every VPN, firewall, and gateway like it’s already under attack: patch fast, enable strict access controls, and send those logs to something that actually gets looked at. Second, if you’re in logistics, telecom, or AI, assume you’re on a shopping list in Beijing; lock down code repositories, MLOps consoles, and any exposed admin panels. Third, follow CISA and FBI alerts in real time, and rehearse an incident‑response plan that assumes a China‑nexus actor moves in under half an hour. I’m Ting, keeping an eye on the digital dragon so you don’t have to. Thanks for tuning in, and don’t forget to subscribe. This has been a quiet please production, for more check out quiet please dot ai. For more http://www.quietplease.ai Get the best deals https://amzn.to/3ODvOta

Dragons in the Banquet Hall: China's Cyber Spies Are Hiding in Your Trusted Software and Nobody Saw It Coming

This is your Digital Dragon Watch: Weekly China Cyber Alert podcast. I’m Ting, and this week’s Digital Dragon Watch is all about how China’s cyber scene keeps mixing stealth, scale, and speed. The biggest verified China-related story in the last few days is the continued fallout from the 2025–2026 wave of state-linked intrusions, with *TechCrunch* reporting that Chinese spies were accused in a breach involving the cybersecurity newsletter authors behind “This Week in Security,” a reminder that even the people tracking the hacks can become targets[1]. What matters most right now is the attack pattern. According to *TechCrunch*, one of the broader 2026 trends is the shift toward quieter, more persistent operations against civilian infrastructure, open-source software, and identity-rich systems rather than splashy one-and-done break-ins[1]. That matters for China-linked risk because the same playbook has been seen in recent campaigns against cloud services, developers, and organizations that manage sensitive data. The new attack vectors that security teams are watching include supply-chain compromise, credential theft from developer tooling, and abuses of legitimate software already trusted inside networks[1][11]. For targeted sectors, the list is broad but very practical: software developers, technology vendors, government systems, and infrastructure operators are all in the blast radius. *TechCrunch* notes that attacks on power, water, and other civilian services have become a troubling pattern, while open-source projects such as Trivy, Bitwarden, and Checkmarx were compromised in separate incidents this year, showing how attackers can reach downstream victims through trusted code paths[1]. That is the cyber version of hiding a dragon in the banquet hall. On the U.S. government side, the clearest official move remains the Federal Bureau of Investigation’s escalation of major cyber incident handling earlier this year, which *TechCrunch* says triggered legally required congressional disclosure after a surveillance system was compromised[1]. While that specific case was not framed as China-linked in the source, it shows the U.S. response posture: treat espionage-grade intrusions as national security events, not routine IT incidents[1]. In practice, U.S. agencies have been pressing for faster reporting, tighter identity protection, and better supply-chain defenses across critical sectors. Expert recommendations are consistent and very concrete. Huntress emphasizes defending against “living off the land” attacks, where intruders use legitimate tools instead of obvious malware, so organizations need stronger logging, anomaly detection, and privilege control[11]. *TechCrunch* also points to the need for basic hygiene that still gets missed: patch faster, restrict developer tokens, segment sensitive systems, and protect government-issued identity documents because exposed passport and license scans can be weaponized for fraud and persistence[1]. So the big takeaway, listeners, is this: China-related cyber risk is less about fireworks and more about invisible footholds, trusted software, and patient espionage. If your team is not watching code-signing, identity exposure, and admin-tool abuse, you are basically leaving the side gate open. Thanks for tuning in, and remember to subscribe. This has been a quiet please production, for more check out quiet please dot ai. For more http://www.quietplease.ai Get the best deals https://amzn.to/3ODvOta

Dragon Routers and AI Phishing: Why Beijing Wants Your Boring Water Utility More Than Your Secrets

This is your Digital Dragon Watch: Weekly China Cyber Alert podcast. Hey listeners, Ting here with your Digital Dragon Watch, and this week the dragon’s been poking at a lot of keyboards. Let’s start with the headline act: the US–China cyber tug‑of‑war over critical infrastructure. In the last few days, US officials have been name‑checking Volt Typhoon, the China‑nexus group that the FBI and CISA say has been quietly burrowing into power, water, and telecom networks across the United States, not just on military bases but in places like Hawaii and Guam. According to public CISA and FBI joint advisories, the new twist is their heavy use of living‑off‑the‑land tools and compromised small office routers, so your average home Netgear becomes a beachhead instead of some Hollywood‑style malware launcher. Targeted sectors? Think boring but vital: electric utilities, regional ISPs, municipal water, and transportation systems. The dragon isn’t trying to steal your Netflix password; it is pre‑positioning for potential disruption in a crisis over places like Taiwan or the South China Sea, a point the Office of the Director of National Intelligence has been hammering in recent worldwide threat assessments. On the US response side, the Justice Department and FBI have been bragging about remote operations to neuter China‑controlled botnets built on those compromised routers, while CISA has been pushing new Shields Up–style guidance aimed specifically at state and local infrastructure operators. The White House’s cyber team has been signaling that Chinese targeting of critical infrastructure now sits alongside Russian ransomware as a top‑tier national security risk, not just an IT problem for your local utility’s help desk. Over in the espionage lane, Microsoft and Google security teams have reported ongoing China‑linked campaigns against US defense contractors and think tanks, with AI‑generated phishing lures getting sharper. Instead of “urgent invoice,” listeners are seeing invites to real‑world conferences or documents that mention actual Hill staffers and committee names. That’s a big evolution in attack vectors: social engineering now tuned by large language models, plus cookie theft and OAuth abuse to bypass your shiny MFA. Defense isn’t standing still. CISA, NSA’s Cybersecurity Collaboration Center, and industry ISACs have been pushing configuration baselines that kill a lot of these tricks: disabling legacy protocols, tightening token lifetimes, enforcing phishing‑resistant MFA like FIDO keys, and segmenting OT networks from corporate IT so a phished marketing intern can’t turn off the lights in Phoenix. So what should you do if you’re not running a power grid but you do not want to be an accidental node in Beijing’s next botnet? Experts from places like Mandiant and CrowdStrike keep repeating the same greatest hits: patch edge devices ruthlessly, turn on hardware‑key MFA for admins, monitor for weird PowerShell and WMI abuse, and log everything to something you actually look at. For smaller orgs, follow CISA’s “secure by design” and “secure by default” guidance, lean on managed detection if you don’t have a 24/7 SOC, and practice incident response like it’s a fire drill, not a board presentation. I’m Ting, and that’s your Digital Dragon Watch for this week. Thanks for tuning in, and don’t forget to subscribe so the next China cyber plot twist finds you before it finds your network. This has been a quiet please production, for more check out quiet please dot ai. For more http://www.quietplease.ai Get the best deals https://amzn.to/3ODvOta

LinkedIn Lures and Cloud Ghosts: Why Chinese Spies Are Sliding Into Your DMs With Fake Job Offers

This is your Digital Dragon Watch: Weekly China Cyber Alert podcast. Hey listeners, Ting here with your Digital Dragon Watch, and the last week has been spicy in China cyber land, so let’s jack in. Let’s start with the fresh joint advisory from the FBI, MI5, and the governments of Australia, Canada, and New Zealand, as reported by TechCrunch. According to that advisory, Chinese intelligence officers are leaning hard on LinkedIn and other job platforms, masquerading as recruiters for fake overseas companies. They are targeting Western professionals with access to non‑public data, especially security‑cleared personnel, Indo‑Pacific military staff, defense contractors, journalists, academics, and think‑tank analysts. The vector isn’t malware; it’s psychology. The playbook is slow‑burn relationship building: flattery, “consulting” offers, and then the quiet ask for sensitive insights. The advisory amounts to a public warning shot from the Five Eyes, telling both government and private sector: treat unsolicited recruiter outreach as a potential intelligence operation, not a networking opportunity. While that’s happening in the open web, in the shadows we’ve got campaigns like Operation Dragon Weave, detailed by researchers at Hexnode. This one is a China‑linked espionage operation hitting organizations in the Czech Republic and Taiwan, especially government, public services, research, academia, tech, and financial services. The attackers kick things off with convincing spear‑phishing emails, often themed around things like Czech Social Security meetings, and pack ZIP attachments that drop Rust‑based malware dubbed Rustcloak. For command‑and‑control, they use an agent called Azureveil that hides traffic in Microsoft Azure Blob Storage, blending in with normal cloud noise. That’s classic “living in the cloud” tradecraft: no sketchy servers, just abusing trusted infrastructure. On the financially motivated side, threat‑intel from SOC Prime highlights a Chinese‑speaking group known as TA4922. They are running credential‑phishing campaigns using HR, payroll, tax, and invoicing lures to trick employees into surrendering login data. Their targets are broad across corporate environments, but the theme is consistent: weaponize everyday business paperwork to punch through the front door. So how are defenders responding? U.S. and allied agencies in the Five Eyes advisory push specific recommendations: verify recruiter identities through official channels, route any approach that touches on sensitive topics to security officers, and train staff that “side gigs” with unknown firms are a risk surface, not a perk. Cloud security experts analyzing Dragon Weave stress deeper inspection of traffic to services like Azure, strict identity and access controls, and threat hunting for odd patterns in Blob Storage use. Email security teams are doubling down on phishing‑resistant authentication, attachment sandboxing, and user reporting drills. And across the board, experts recommend continuous monitoring for living‑off‑the‑land behavior: trusted tools or platforms doing very untrusted things. I’ll leave you with this: the most dangerous exploit right now isn’t a zero‑day, it’s a zero‑skepticism professional on LinkedIn and a cloud tenant nobody’s watching closely. Thanks for tuning in, and don’t forget to subscribe. This has been a quiet please production, for more check out quiet please dot ai. For more http://www.quietplease.ai Get the best deals https://amzn.to/3ODvOta

China's Cloudy with a Chance of Espionage: Azure Blobs, Rust Loaders, and Why Your LNK Files Need Therapy

This is your Digital Dragon Watch: Weekly China Cyber Alert podcast. I’m Ting, and this week’s China cyber weather report is a little stormy: the clearest fresh campaign is **Operation Dragon Weave**, a China-linked espionage operation that used LNK shortcut lures, a Rust loader, and Microsoft Azure Blob Storage as command-and-control to hit government personnel and researchers in **Taiwan** and **Czechia**. According to **SOC Prime**, the chain moved from a ZIP file to VBScript, PowerShell decryption, DLL sideloading, and a custom Rust loader that decrypted the final payload with RC4, Base64, and SM4, which is a very polished way to say “quietly very annoying.” The standout new attack vector is the abuse of **cloud storage as C2**, especially Azure Blob Storage, because it blends in with ordinary enterprise traffic and makes takedown harder. **SOC Prime** says the last-stage malware, called **AZUREVEIL**, communicates only through Azure Blob Storage and can execute Beacon Object File payloads in memory, which is the sort of detail defenders want before the coffee gets cold. The targeted sectors in this campaign are **government** and **research**, especially people handling sensitive regional policy, technical analysis, or cross-border intelligence. On the defensive side, the lesson is blunt: treat **LNK files, ZIP attachments, and script launch chains** as high-risk, especially when they trigger wscript, PowerShell, or unusual DLL sideloading. SOC Prime’s reporting implies defenders should hunt for multi-stage behavior, not just one malicious hash, because the attack survives by chaining normal-looking tools together. In practice, that means tightening endpoint rules, restricting script interpreters, watching for suspicious Azure storage access, and correlating file execution with network beacons. Now, zooming out to the broader China-related threat picture for the past week, the most important pattern is that espionage crews are increasingly using **living-off-the-land** techniques and cloud infrastructure to blend into legitimate traffic. That matters because the old “block the bad IP” playbook is not enough when the attacker is hiding inside Microsoft Azure or borrowing trusted Windows components. For official U.S. government response, the strongest directly relevant recent move in the available reporting is the White House’s new framework to **vet top AI models for national security risks**, which reflects Washington’s growing concern that advanced AI can amplify cyber operations, even if that order is not China-specific in the narrow sense. That kind of policy signal matters because cyber defenders are now worrying not only about malware, but about AI-assisted reconnaissance, phishing, and automation. Expert recommendations are consistent across the current threat landscape: reduce reliance on static indicators, monitor for **multi-step intrusion chains**, segment high-value research and government networks, and make sure cloud logs are actually being reviewed rather than admired from a distance. If I had to say it in one sentence, listeners: the new China cyber playbook is less smash-and-grab and more stealth, cloud, and patience. Thank you for tuning in, subscribe for more, and this has been a quiet please production, for more check out quiet please dot ai. For more http://www.quietplease.ai Get the best deals https://amzn.to/3ODvOta