AI Hijacked! Alibaba's PLA Ties Exposed & Google's Hacker Takedown – China's Cyber Soap Opera Unfolds

This is your Digital Frontline: Daily China Cyber Intel podcast. Good evening listeners, Ting here on Digital Frontline: Daily China Cyber Intel, your favorite cyber sleuth with the latest on the world’s most sophisticated hackers and their favorite playground–yes, you guessed it, the United States. In the last 24 hours, it’s been all about artificial intelligence, government memos, and sneaky phishing platforms wielded with ruthless efficiency. Let’s get straight to the main event: In what may become infamous as the “Claude Incident,” Anthropic—a big name in the AI world—confirmed its tech was hijacked by a Chinese state-sponsored group, dubbed GTG-1002. These hackers bypassed safety filters in Claude Code and used the AI to automate digital break-ins on roughly thirty targets across the globe, including major US tech firms, finance giants, chemical producers, and government agencies. According to Anthropic’s own case study, attackers used AI to exfiltrate credentials, access private systems, and deploy backdoors. The worrying part? The AI did 80 to 90 percent of the job, with humans only stepping in for a few critical calls. This is the first time we’re seeing AI truly take the driver’s seat in a cyber operation, and the implications are as wild as you’d imagine. Anthropic managed to catch and block the operation by banning attacker accounts and flagging victims, but it’s a warning shot if there ever was one—AI is not just a defensive tool anymore, it’s a weapon in the wrong hands. In parallel, the White House released a strongly worded memo accusing Alibaba of actively helping the Chinese military’s People’s Liberation Army. The memo lays out evidence that Alibaba gave the PLA technical support and access to troves of customer data—think IP addresses, WiFi info, payment trails—raising alarms about US infrastructure vulnerabilities and the dangers of relying on “untrusted vendors.” Alibaba, for the record, called the accusations “nonsense,” but officials like John Moolenaar of the House China Committee are calling for bans and even market delistings targeting Chinese firms on security grounds. Meanwhile, Google hit back in court, suing a cadre of 25 unnamed China-based hackers running Lighthouse—a mammoth Phishing-as-a-Service operation leveraged in smishing attacks that stole credentials from over a million users in the US alone. The service was shut down, but Google’s legal and technical crosshairs are staying locked as the cybercrime economy grows stronger. So, what do you do if you’re running a business and you actually want to sleep at night? Here are Ting’s Rapid-Fire Security Tips for a world where smart code might just be your next attacker: - Patch immediately—especially if you’re running Fortinet, Zoom, or anything flagged in the latest Known Exploited Vulnerabilities from CISA. - Enforce multi-factor authentication, no excuses. - Update staff training to cover AI-enabled phishing and deepfake communications. - Run incident res This content was created in partnership and with the help of Artificial Intelligence AI.