Report: DeepSeek and Moonshot AI used OpenAI model outputs to train their latest models

Research Buddy — Today’s News & Updates

Thursday, 29 October 2026 · India, technology, business and employment

Today’s briefing looks at the developments shaping India’s job market, enterprise AI adoption, the global AI race and the skills employers may need next. Each story includes the context behind the headline and why it matters.

5. Report: DeepSeek and Moonshot AI used OpenAI model outputs to train their latest models

Reuters, citing The Information · 29 October 2026

A report by The Information, cited by Reuters, says Chinese AI developers DeepSeek and Moonshot AI relied heavily on outputs from OpenAI’s GPT-6 Astra to train their latest models. The report is based on people with direct knowledge of the matter. It does not specify exactly how the outputs were used or identify the models involved.

The story touches on a major technical and commercial issue in AI: synthetic training data. AI-generated examples can help developers train or improve models, particularly when suitable human-created data is scarce, expensive or difficult to obtain.

However, using outputs from another company’s model can raise questions about licensing, terms of service, intellectual property and competitive advantage. The specific facts and any potential policy implications in this case remain subject to further reporting.

Why it matters: The AI race is not only about building larger models. It is also about data quality, training techniques, model evaluation and the rules governing how one developer can use another developer’s outputs.

For AI professionals, this reinforces the importance of understanding data provenance, responsible model development and the difference between generating synthetic data and evaluating whether that data improves a model.

What to watch next: Whether the companies or OpenAI comment further, whether more technical details emerge, and whether the report leads to changes in how AI-generated training data is governed.

Source: Read the Reuters report

Editorial source note

This is a reported claim attributed to The Information via Reuters, not independently verified by Aiexl. Direct access to the supplied Reuters link returned HTTP 401 during checking. The supplied GPT-6 Astra reference is retained as provided and is not independently confirmed here. The photograph is illustrative.

Today’s key takeaways

Employment: AI-related business is growing, but hiring remains uneven between companies and roles. GCCs are another important part of India's technology employment landscape.

AI in finance: The securities regulator's proposal illustrates how AI may be applied to oversight and market monitoring, with governance and human review remaining important.

AI development: The reported use of model-generated outputs for training highlights the growing importance of data quality, provenance and responsible development.

Education and careers: Degrees remain valuable, but students also need practical evidence of their skills, including projects, technical explanations and problem-solving.

The big picture for 29 October: AI is moving deeper into enterprise operations, financial supervision and the business models of technology companies. At the same time, the employment benefits are not automatic. The central question for businesses is whether AI delivers measurable value; for workers and students, it is how to develop skills that translate into useful work.