Anthropic accuses Chinese AI firms of data theft
ArchivedGlobal|5 articles|5 sources|4 blocks
Detected: Updated: Analyzed:
NarrativeRadar Analysis·Reviewed by M. Reyes·AI-assisted, editorially supervised·Based on 5 articles from 5 sources
On 2026-02-26, OpenAI said Chinese law enforcement tried to use ChatGPT to plan an influence operation targeting Japan’s prime minister, adding to earlier US claims that Chinese actors are abusing Western AI tools. Days earlier, Anthropic and OpenAI accused several Chinese AI labs of running large-scale “distillation” campaigns to copy their models’ outputs and training data. The dispute raises stakes for US–China tech competition, export controls, and how governments police cross-border use of powerful AI systems.
Reported Facts
Observable data points shared across all narratives
- •Anthropic said on 2026-02-24 that Chinese AI labs carried out 'industrial-scale' distillation attacks on its models.
- •Anthropic described the activity as repeated automated querying of its systems to reproduce its model behavior inside rival Chinese models.
- •OpenAI has separately reported that Chinese-linked actors used its tools to support online influence operations, including content about foreign political leaders.
- •One OpenAI disclosure said Chinese law enforcement officials tried to use ChatGPT to plan an influence campaign against Japan’s Prime Minister Fumio Kishida.
- •Reports say both Anthropic and OpenAI have begun blocking or limiting access from accounts and infrastructure they link to these Chinese campaigns.
- •Media accounts describe the Chinese activity as targeting not only Anthropic but also OpenAI’s models, suggesting a broader pattern rather than a single incident.
- •The accusations have prompted discussion in the US and Europe about tightening export controls and access rules for advanced AI models used from China.
- •Russian coverage has repeated Anthropic’s accusation that Chinese AI companies engaged in data theft, highlighting tensions between US and Chinese tech firms.
Core Disagreement— Main Problem
According to Finance, chinese labs copying us models and misusing ai tools. However, China sources see it as foreign ai tools involved in sensitive political activity.
Narrative Split
How different information blocks interpret these facts
CN
Security Threat Framing
Chinese regional coverage focuses on OpenAI’s claim that Chinese law enforcement sought to use ChatGPT for an influence operation against Japan’s leader, presenting it mainly as a security and foreign policy story rather than a business dispute. The reporting highlights how foreign AI tools can be used in information campaigns but gives little detail on the accused Chinese AI labs or their response. Commentators suggest the episode will fuel more controls on foreign AI platforms and greater push for homegrown Chinese systems.
- •OpenAI reports that Chinese law enforcement personnel attempted to use ChatGPT to plan an influence operation targeting Japan’s Prime Minister Fumio Kishida.
- •Chinese discussion stresses the risk that foreign AI platforms can be used in sensitive political work.
- •Coverage gives limited attention to Anthropic’s specific accusations against Chinese commercial AI labs.
- •Chinese voices predict that foreign restrictions will push China to speed up development of domestic AI models.
- •Officials and experts in China are expected to call for tighter management of how foreign AI tools are used inside the country.
RU
US–China Tech Rivalry
Russian outlets describe Anthropic’s accusation as another front in US–China technology rivalry, with American firms accusing Chinese companies of theft while both sides race to build powerful AI. They present the dispute as part of a broader struggle over who controls key digital tools and standards, rather than a narrow legal case. Commentators in this block expect more mutual restrictions, with Russia watching for chances to deepen ties with Chinese tech firms that face Western limits.
- •Russian coverage reports that Anthropic accused Chinese AI companies of stealing data from its models.
- •Russian commentators link the accusation to a wider US effort to slow China’s rise in advanced technologies.
FINANCE
IP Theft Concerns
Financial and Western outlets present Anthropic’s and OpenAI’s claims as evidence that Chinese AI labs and state-linked actors are trying to shortcut development by copying US models and using them for political influence. They stress the risks to intellectual property, national security, and investor confidence in AI firms whose products can be quietly copied or misused from abroad. Commentators expect pressure for tighter export controls, stricter access screening, and clearer rules on what counts as illegal model copying.
- •Anthropic accuses Chinese AI labs of running automated distillation attacks to clone its models’ behavior.
- •OpenAI reports that Chinese law enforcement tried to use ChatGPT to design an influence campaign against Japan’s prime minister.
Create a free account to read all perspectives
Already have an account? Sign in
Where Narratives Diverge
Key disagreements, blind spots, and what to watch next.
Main Problem◇Different Reading
Finance
Chinese labs copying US models and misusing AI toolsChina
Foreign AI tools involved in sensitive political activitySo what
Readers cannot easily tell whether the core issue is theft of US technology or the political use of foreign AI platforms.
US Motive◇Different Reading
Finance
US firms defending intellectual property and securityRussia
US using theft claims to slow Chinese tech growthSo what
It is hard to judge whether new restrictions are mainly about fair competition or about keeping China behind.
Chinese Role⚡Disputed
Finance
Chinese labs and law enforcement directly linked to campaignsChina
Reports mention law enforcement but not specific lab wrongdoingSo what
Without clear naming of the Chinese labs involved, readers cannot know how broad or organized the alleged copying effort is.
Legal Standards○Nobody Covers
None of the blocks clearly explain which specific US or Chinese laws the alleged distillation attacks might violate, making it hard to see whether this is a criminal matter, a civil dispute, or mainly a political fight.
Government Action▸What to Watch
If US or Chinese regulators open formal investigations or announce sanctions against named AI labs in the next few months, that would show whether authorities treat the accusations as serious legal violations rather than just corporate complaints.
What Could Happen If...
- ▸If US regulators treat Anthropic’s and OpenAI’s reports as evidence of illegal data theft by Chinese labs Washington could tighten export controls on advanced AI models and restrict access from Chinese IP ranges, affecting Chinese researchers and companies that rely on US tools.
Market Exposure
Different sides disagree on how this affects markets. The same instrument may move in opposite directions depending on which reading proves correct.
According to Finance sources
StocksNVIDIA CorporationIncreased Volatility
If US–China AI tensions lead to tighter export controls on advanced chips and AI services, demand from Chinese customers for NVIDIA hardware could be restricted while US and allied demand for secure AI infrastructure grows, pulling its share price in opposite directions.
Key Actors
Organizations
Source Coverage
RU
1 articles
CN
1 articles
FINANCE
2 articles
REGIONAL
1 articles
More coverage on this story
- Topic: US–China Economic Conflict
- More on People's Republic of China
- More on United States of America
- More on Anthropic
Narrative Analysis
Divergence Detectedconfidence: 95%
Signals
Cross-block coverage detectedMultiple blocks covering the same event
Related Events
▸Trump Administration Targets Chinese Firms Over U.S. AI Model UseArchived▸U.S. Accuses China of Large-Scale AI Technology TheftArchived▸China is leading in holistic AI development. Can the US catch up?Archived▸Christopher Wood identifies Anthropic as key AI market playerArchived▸China Advances in AI Race Amid Global Industry Shifts
Primary Sources (5)
RUKommersant (RU)Anthropic обвинил китайские ИИ-компании в краже данныхCNChannel News AsiaChinese law enforcement tried to use ChatGPT to plan influence op against Japan PM: OpenAIFINANCECNBC - Top NewsAnthropic joins OpenAI in flagging 'industrial-scale' distillation campaigns by Chinese AI firmsFINANCEFinancial Times - WorldAnthropic accuses Chinese AI labs of ‘distillation attacks’ on its models