TECH

Alibaba Bans Staff From Using Anthropic AI Tools Over Security Concerns

Published: Jul. 3, 2026 11:16 p.m. GMT+8

00:00

00:00/00:00

Listen to this article 1x

Alibaba will require employees to stop using Claude Code and has added it to a high-risk software list, an insider confirmed to Caixin on July 3, citing backdoor risks. Photo: VCG

Alibaba Group Holding Ltd. is barring employees from using Anthropic’s artificial-intelligence tools starting July 10, citing security concerns after the U.S. startup acknowledged it had embedded hidden code to track certain users.

An Alibaba insider confirmed to Caixin on Friday that the Chinese e-commerce giant has added Anthropic’s Claude Code to its high-risk software list. Employees will be required to uninstall Anthropic models and agent products and switch to Alibaba’s in-house AI assistant, Qoder, the person said.

You've accessed an article available only to subscribers

Subscribe today for just $.99.

VIEW OPTIONS

Unlock exclusive discounts with a Caixin group subscription — ideal for teams and organizations.

Save an extra $50. Introductory offer for new readers. Subscribe now.

Share this article

Open WeChat and scan the QR code

DIGEST HUB

Digest Hub

Back

Explore the story in 30 seconds

Alibaba will ban employees from using Anthropic's AI tools from July 10 over hidden tracking code concerns.
Anthropic said the code was a test mechanism to prevent abuse and has been removed in an updated version.
The ban follows Anthropic's accusation that Alibaba conducted a large-scale distillation attack using 25,000 fraudulent accounts.

AI generated, for reference only

Who’s Who

Alibaba Group Holding Ltd.: Alibaba Group banned employees from using Anthropic’s AI tools, citing security risks from hidden tracking code. Employees must uninstall Anthropic models and switch to Alibaba’s in-house AI assistant, Qoder. The move escalates tensions over alleged data theft and model distillation accusations between the two companies.

Anthropic: Anthropic is a U.S. AI startup that barred mainland China and Hong Kong users from its tools. It embedded hidden tracking code in Claude Code to prevent abuse, but plans to remove it. The company also accused Alibaba of a major distillation attack and now requires identity verification for Claude users.

AI generated, for reference only

What Happened When

July 2024：: Anthropic barred users in the Chinese mainland and Hong Kong from directly accessing its AI tools.

September 2025：: Anthropic expanded restrictions to include any company or subsidiary globally that is majority-owned by a Chinese entity.

March 2026：: Anthropic launched a test mechanism involving hidden tracking code in Claude Code to prevent unauthorized intermediaries from abusing accounts.

April 22, 2026：: Start of period in which operators linked to Alibaba, its Qwen LLM, and affiliated AI labs allegedly conducted interactions with Anthropic models using fraudulent accounts.

June 5, 2026：: End of period in which operators linked to Alibaba, its Qwen LLM, and affiliated AI labs allegedly conducted 28.8 million interactions with Anthropic models using about 25,000 fraudulent accounts.

June 24, 2026：: CNBC reported that Anthropic sent a letter to the U.S. Senate accusing Alibaba of carrying out 'the largest distillation attack against Anthropic to date.'

June 30, 2026：: A Reddit post alleged that Anthropic had embedded hidden tracking code in Claude Code, collecting data on Chinese users without their knowledge.

July 3, 2026：: An Alibaba insider confirmed that the company added Anthropic's Claude Code to its high-risk software list and will require employees to uninstall Anthropic models and switch to Alibaba's in-house AI assistant, Qoder. As of publication, Anthropic and Alibaba had not responded to Caixin's requests for comment.

By publication on July 3, 2026：: Thariq Shihipar, an Anthropic researcher, stated that the research team had already planned to remove the tracking mechanism, noting that a code merge request had been completed and the feature would be fully rolled back in an updated version.

AI generated, for reference only