Exclusive: DeepSeek and Moonshot AI relied heavily on outputs from OpenAI’s GPT-6 Astra to train their latest models. The shift comes as Anthropic cracks down on Chinese labs extracting capabilities from Claude. Full story:Read original on X
StoryFirst seen
DeepSeek and Moonshot AI trained models on OpenAI's Astra outputs
The reported use of a rival lab's outputs comes as Anthropic cracks down on Chinese labs extracting capabilities from its models.
Everything it touches
10 names 3 voices 3 source posts
The story’s neighborhood
Explore the graph, or choose a connection below.
Follow the evidence
The source posts
This brief comes from Tech Twitter’s edited headlines. The posts below are the sources grouped with the story; statements in them belong to their authors.
Well, well: Moonshot and DeepSeek are turning to OpenAI's GPT-6 Astra as a distillation target after Anthropic cracked down on such practices.Read original on X
SITUATION DETECTED: Moonshot AI and DeepSeek both distilled OpenAI’s GPT-6 Astra for their latest models. Kimi K3.1 Pro and DeepSeek’s V4.1-Flash both benefited, per The Information.Read original on X
How the headline evolved
DeepSeek and Moonshot AI trained models on OpenAI's Astra outputs
Moonshot AI and DeepSeek reportedly trained on OpenAI's Astra outputs
Moonshot AI and DeepSeek reportedly distilled OpenAI's GPT-6 Astra model
Moonshot AI and DeepSeek reportedly trained on OpenAI's GPT-6 Astra
Moonshot AI and DeepSeek reportedly trained on OpenAI's GPT-6 Astra
Moonshot AI and DeepSeek reportedly trained on OpenAI's GPT-6 Astra outputs
Moonshot AI and DeepSeek reportedly trained on OpenAI's newest model outputs
Moonshot AI and DeepSeek reportedly distilled OpenAI's newest model
Moonshot AI and DeepSeek reportedly trained models on OpenAI's newest model
Moonshot AI and DeepSeek reportedly distilled OpenAI's newest model
Shared connections
Related stories
OpenAI's claimed proof of pure Yang-Mills theory faces public validation
The Yang-Mills Working Group plans to publish OpenAI's submitted code and its validation, so the claim can be checked by mathematicians.
Adobe replaces OpenAI as Disney's generative AI partner
Disney plans to use the Adobe models in its studios and at its theme parks, tailored to its own properties.
Anthropic plans to start trading shares next week
A public listing would give investors direct exposure to one of the largest AI model developers.
Anthropic releases Claude Fable 5.2 and Claude Mythos 5.2 models
Both new models are available now, giving developers and businesses fresh options from a leading AI lab.