Anthropic accuses Alibaba of illicitly accessing its AI

2026-06-25 · Source: Semafor · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy · Depth: Fundamental Awareness, extended

Summary

Anthropic has accused Chinese e-commerce giant Alibaba of illicitly accessing its Claude AI model, marking a significant development in the ongoing US-China tech rivalry. Anthropic alleges that Alibaba created fake accounts to circumvent access restrictions and conduct "distillation attacks," a process where Claude's generated responses are used to train Alibaba's proprietary AI models. This accusation follows Alibaba's recent lawsuit against the Pentagon, seeking removal from a blacklist of firms allegedly linked to the PLA. The broader geopolitical context involves the US's "Pax Silica" initiative, aimed at reducing reliance on Chinese tech supply chains, an effort now supported by several European governments and the EU.

Key takeaway

For Directors of AI/ML and Legal Professionals, this incident underscores the escalating risk of intellectual property theft in AI. You must urgently review your AI model access controls and terms of service to prevent "distillation attacks" and unauthorized training. Be aware that geopolitical tensions are amplifying these threats, necessitating robust legal and technical safeguards to protect your proprietary models from illicit competitive exploitation.

Key insights

AI model access controls are vulnerable to circumvention for competitive training via "distillation attacks".

Principles

• AI model security requires robust access controls.
• Geopolitical rivalries intensify tech IP risks.
• "Distillation attacks" exploit model outputs for training.

Method

Illicitly accessing a competitor's AI model through fake accounts to generate responses, then using these responses to train one's own models.

In practice

• Strengthen AI model access authentication.
• Actively monitor user behavior for anomalies.
• Review terms of service for AI model usage.

Topics

• AI Intellectual Property
• AI Model Security
• Distillation Attacks
• US-China Tech Rivalry
• Geopolitical Risk
• Supply Chain Security

Best for: CTO, VP of Engineering/Data, Executive, Director of AI/ML, Legal Professional, Tech Journalist

Related on AIssential

• Google and OpenAI complain about distillation attacks that clone their AI models on the cheap — AI model distillation poses a significant intellectual property theft risk, enabling cheap cloning by extracting reasoning steps.
• Amazon CEO reportedly raised Anthropic model concerns before government crackdown — Amazon's security concerns about Anthropic's AI models led to a government export ban and model de-deployment.
• Anthropic says Alibaba must be punished for largest Claude cloning attack — Chinese firms are allegedly conducting large-scale distillation attacks to replicate US frontier AI capabilities, raising national security concerns.
• Anthropic slams Chinese AI firms for harvesting data from its Claude chatbot — AI frontier models accessed via APIs present a new attack surface for industrialized data distillation.
• Detecting and preventing distillation attacks — Illicit AI model distillation poses significant national security risks and undermines export controls.
• Are Anthropic’s New Claude Models Evading Human Control? — The rapid recall of Anthropic's advanced AI models due to automated hacking risks highlights the immediate challenge of controlling powerful AI.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Semafor.