Anthropic accuses DeepSeek and other Chinese firms of using Claude to train their AI

Anthropic claims DeepSeek and 2 different Chinese AI companies misused its Claude AI exemplary successful an effort to amended their ain products. In an announcement connected Monday, Anthropic says the "industrial-scale campaigns" progressive the instauration of astir 24,000 fraudulent accounts and much than 16 cardinal exchanges with Claude, arsenic reported earlier by The Wall Street Journal.

The 3 companies - DeepSeek, MiniMax, and Moonshot - are accused of "distilling" Claude, oregon grooming a smaller AI exemplary based connected a much precocious one. Though Anthropic says that distillation is simply a "legitimate grooming method," it adds that it tin "also beryllium utilized for illicit intent …

Read the afloat communicative astatine The Verge.