Alibaba Bans Claude Products Internally
Alibaba has announced a comprehensive ban on the Claude product series from Anthropic, requiring all employees to uninstall Agent tools including Sonnet, Opus, Fable, and Claude Code.
The ban will officially take effect on July 10, covering multiple models and development tools, aimed at strengthening data security and compliance management.
This move comes after Anthropic accused Alibaba affiliates of extracting Claude capabilities through a large number of fake accounts, highlighting the escalating friction in AI technology between China and the U.S.
Source: Public Information
ABAB AI Insight
Anthropic previously accused Alibaba-affiliated Qwen Lab of using nearly 25,000 fake accounts for 28.8 million interactions to conduct distillation attacks on Claude's core capabilities, exacerbating the trust crisis between the two parties.
Alibaba's internal ban cuts off potential data leakage paths while promoting the replacement with its own models, concentrating capital and resources on domestic AI infrastructure to reduce external dependency risks.
Similar to historical corporate responses to supply chain security incidents, Alibaba is accelerating the construction of an independent technology stack in the context of U.S.-China AI decoupling, to address export controls and accusations.
Structural judgment: Essentially a shift towards technological substitution. Regulatory friction and security accusations are driving companies to move from reliance on foreign cutting-edge models to self-developed alternatives, restructuring the AI toolchain for local control and data sovereignty.
ABAB News · Cognitive Law
Technological dependence is a risk exposure, and the ban accelerates the replacement loop.
Friction forces local innovation, with replacement costs lower than long-term leakage.
When security is prioritized, external tools give way to internal control.