Alibaba has ordered staff to stop using Claude Code by July 10, after a Reddit user's reverse engineering exposed hidden code that checked whether users were tied to China.
If you're an engineer at Alibaba, you have until July 10 to find a new coding assistant. The company has added Anthropic's Claude Code to its internal list of high risk software, the category normally reserved for tools with known security vulnerabilities, and told staff to switch to Qoder, its own in house coding platform, instead.
The ban traces back to a Reddit post from June 30. A user going by LegitMichel777 published a breakdown of Claude Code's source code, claiming that since version 2.1.91, released April 2, the tool quietly checked a user's proxy settings and system time zone against two hidden lists. According to reporting from Cybernews and Techzine, one list contained 147 domains tied to Chinese companies and AI labs, Baidu, Alibaba, Ant Group and ByteDance among them, alongside eleven keywords linked to labs including Moonshot AI. The result of that check got folded into an ordinary looking string in the code, 'Today's date is...', where a hyphen swapped for a slash flagged a Chinese time zone.
That's an odd place to hide a fingerprinting routine.
An Anthropic engineer confirmed the code was real and said on social media that it would be pulled in the next day's release. The company hasn't issued a full written explanation of why the check existed, but its internal framing has been anti-fraud, not surveillance.
The tracking code didn't appear out of nowhere. In a June 10 letter Anthropic sent to U.S. senators, the company accused operators tied to Alibaba's Qwen lab of running close to 25,000 fraudulent accounts that generated more than 28.8 million interactions with Claude between April 22 and June 5. Anthropic's argument was that the accounts existed to distill Claude's outputs and speed up training of a rival model. Seen against that letter, the hidden check looks less like espionage and more like Anthropic trying to identify the exact customers it had already accused of gaming its system.
That irony is hard to miss. Claude Code was never officially sold in mainland China, Anthropic blocks direct sign ups from Chinese IP addresses, which meant most Chinese developers using it were already routing around that restriction through VPNs or third party resellers, exactly the kind of access Anthropic says its hidden list was built to catch in the first place.
Alibaba isn't buying that framing. According to Caixin Global's report on the internal notice, the company told staff plainly to stop using Claude Code for work, then pointed them to Qoder, the coding assistant it has spent the year building out. The South China Morning Post reported that Alibaba's security team classified Claude Code as carrying 'back door risks,' language usually kept for compromised software, not a competitor's product.
The timing is awkward for both sides. Alibaba's Qwen models compete directly with Claude for developer attention, and the fraudulent traffic Anthropic flagged in its Senate letter was, by Anthropic's own account, an attempt to shortcut that competition through distillation. Now Alibaba gets to cast itself as the victim of an American company's surveillance code, a far easier story to tell staff than admitting its affiliated developers were caught scraping a rival's model.
Neither company comes out of this looking careful.
Anthropic says the tracking mechanism is already gone or going in its next release, but a patch note doesn't undo a Reddit post that's been read by tens of thousands of developers. Alibaba's staff are bound to Qoder starting next week, and Baidu and ByteDance, both named in that same hidden list, are watching how this settles before deciding whether they need bans of their own.
Also read: Israel Turns Its Pact With Argentina Into an AI Foothold in Latin America • Quantum Systems Raises $1.2 Billion as Airbus Bets on a Future Rival • China's AI-Driven Quant Funds Have Now Topped $474 Billion in Assets
