DeepSeek and Moonshot AI Used OpenAI's GPT-6 Astra Outputs to Train Models
Chinese AI firms DeepSeek and Moonshot AI have extensively used outputs from OpenAI's GPT-6 Astra model to train their latest AI models, a practice known as distillation. This method, which accelerates and reduces the cost of replicating advanced AI capabilities, has become contentious amid US-China tech competition. Moonshot's Kimi K3.1 Pro and DeepSeek's V4.1-Flash 1017 models benefited from this approach. OpenAI and other companies like Anthropic have taken measures to limit data harvesting through distillation attacks.
First-hand measurement across 2 sources
We measured how 2 outlets covered this story. No outlet gave this story a measurable political slant — there is no left–right reading to report. Overall sentiment is neutral (50/100). Lens Score 43/100.
Outlets measured: ndtv, economictimes. See how each one headlined and framed the same story in the source comparison below.
AI Analysis
Sentiment was consistent across outlets (50–50/100), indicating broadly factual reporting rather than editorialising.
Coverage timeline
economictimes broke this story on 29 Oct, 02:48 am. Other outlets followed.
- 1