Are China’s ‘AI tigers’ cheating? US rival Anthropic alleges some are

illuminem summarises for you the essential news of the day. Read the full piece on CNN or enjoy below:

🗞️ Driving the news: U.S. AI firm Anthropic has accused three leading Chinese AI labs — DeepSeek, MiniMax and Moonshot AI — of illegally “distilling” its Claude model to advance their own systems
• Anthropic alleges the firms created over 24,000 fraudulent accounts and generated more than 16 million Claude interactions to extract capabilities

🔭 The context: Distillation, training a smaller model using outputs from a more advanced one, is common in AI, but most frontier labs ban third-party extraction
• OpenAI recently made similar allegations against DeepSeek
• The claims come amid intensifying U.S.-China tech rivalry and ongoing debate over the effectiveness of U.S. export controls on advanced AI chips

🌍 Why it matters for the planet: AI competition shapes global energy demand, semiconductor supply chains and digital governance norms
• Large-scale AI deployment significantly increases electricity consumption through data centers
• Geopolitical tensions over AI leadership may influence global cooperation on AI safety, climate modeling and cybersecurity safeguards

⏭️ What’s next: Chinese firms have yet to publicly respond to the latest allegations
• U.S. policymakers may use the claims to justify tighter export controls and AI governance measures

💬 One quote: “The window to act is narrow.” — Anthropic

📈 One stat: Anthropic alleges over 16 million exchanges with Claude were used in the distillation process

See on illuminem's Data Hub™ the sustainability performance — carbon credit purchases, total emissions, and climate targets of thousands of companies

Click for more news covering the latest on AI & green tech