February 25, 2026 Source: SecurityWeek 3 min read · 681 words

Google Disrupts Chinese Hackers Targeting Telecoms, Governments

Google розпорошує китайських хакерів, які атакували телекомунікаційні компанії та уряди 42 країн

Google Disrupts Chinese Hackers Targeting Telecoms, Governments Across 42 Countries

Forty-two countries. Since 2017. And nobody caught it until now.

Google's threat intelligence team just disrupted an active cyberespionage campaign run by Chinese threat actor UNC2814 that's been systematically compromising telecommunications companies and government organizations for nearly a decade. According to SecurityWeek, this isn't some theoretical vulnerability disclosure or a future threat—this is a real, ongoing operation that was actively stealing data when Google pulled the plug.

Let that sink in. A coordinated Chinese cyber attack of this scale, targeting this many critical infrastructure targets across this many nations, and it took this long to disrupt. This is the kind of story that should make every telecom executive and government CISO lose sleep tonight.

Breaking It Down

Here's what we know: UNC2814 has been operating in the shadows since at least 2017, which puts this campaign right up there with some of the most persistent Chinese cyber attacks in recent memory. The targets aren't random—they're strategic. Telecommunications providers are the nervous system of modern infrastructure. Governments run the show. Hitting both simultaneously gives you visibility into who's talking to whom, what they're saying, and where the power flows.

Google's disruption efforts mean the threat actor's primary infrastructure is now offline.

But here's the thing that keeps me up: for nine years, this campaign operated in relative obscurity. Think about what that tells us about detection capabilities. We talk endlessly about china's vulnerability paradox—the idea that China's aggressive posture masks defensive weaknesses—yet Chinese threat actors consistently outmaneuver us on the espionage front. There's a disconnect there.

The real question is whether this disruption actually stops the operation or just forces them to relocate. These aren't amateurs. They've had nearly a decade to build redundancy and backup infrastructure. Google's action is important, but it's not necessarily game-over.

The Technical Side

SecurityWeek didn't release extensive technical details—which is standard practice when you're dealing with active threat actors—but we know UNC2814 was using this campaign for classic espionage objectives. Establishing persistent access. Maintaining it across multiple victim networks. Exfiltrating intelligence without triggering alarms.

This required sophistication.

The attack almost certainly involved credential compromise, lateral movement techniques, and methods designed to evade detection across heterogeneous network environments. Telecommunications networks are notoriously complex—legacy systems running alongside modern infrastructure, multiple jurisdictions, regulatory constraints. An effective campaign here demands attackers who understand the landscape deeply.

And frankly, that level of operational maturity didn't develop overnight. The tools and techniques used in this campaign have probably been refined across hundreds of smaller operations, many of which we'll never see in public disclosure.

Who's Affected

Forty-two countries. That's the headline number. But what matters is specificity: which telecoms are compromised? What government agencies? How much intelligence left the network?

SecurityWeek's reporting focuses on the disruption itself rather than victim attribution—smart editorial decision, since naming targets could expose them further.

If your organization operates in telecommunications or government services across Asia-Pacific, Europe, North America, or the Middle East, you should assume you might be in scope. This wasn't a targeted attack on one nation or sector. It was industrial-scale espionage.

What To Do Now

First: check your logs. If your organization touches telecommunications or government infrastructure, forensic teams should be hunting for signs of UNC2814 indicators of compromise. Google will release these, and so will other threat intelligence vendors. Use them.

Second: assume your environment's been pre-compromised.

Not necessarily by UNC2814, but by something. Organizations targeted at this level typically have multiple threat actors in their networks simultaneously. Disrupting one doesn't clean the others out.

Third: credential rotation isn't optional anymore—it's mandatory. Every sensitive account, every administrative access, every VPN credential tied to potentially compromised systems should be cycled immediately. This is expensive and disruptive. Do it anyway.

Finally, demand access to China's National Vulnerability Database and cross-reference any CVEs that might've enabled initial access here. Understanding the vulnerability disclosure patterns behind campaigns like this one tells you which zero-days matter most for defensive prioritization.

This disruption is real progress. But the campaign existed for nine years undetected. That's the actual story.

Read original article →

// FAQ

Was my telecom company or government agency targeted by UNC2814?

UNC2814 targeted telecommunications companies and government organizations across 42 countries since 2017. Check with your organization's security team or Google's published indicators of compromise to determine if you were compromised. Organizations in critical infrastructure should assume they may be in scope and conduct forensic analysis.

How long has the Chinese cyber attack been happening?

According to SecurityWeek, UNC2814's campaign has been active since at least 2017—nearly nine years—before Google disrupted it in February 2026. The extended timeline shows how long sophisticated Chinese cyber attacks can operate undetected within critical infrastructure.

What should I do if my organization was affected by this attack?

Immediately conduct forensic analysis using indicators of compromise released by Google and threat intelligence vendors, rotate all sensitive credentials, assume pre-compromise across your environment, and engage incident response teams. Contact law enforcement and relevant cybersecurity authorities in your jurisdiction.

Concerned about your project's security? Run an automated pentest with AISEC — AI-powered scanner with expert verification. Go to dashboard →