Launched just two months ago, Discovery Loop has quickly attracted attention in the technology industry thanks to its high-profile founding team and ambitious mission to automate scientific research using artificial intelligence.

The company has reportedly recruited several leading researchers from major US technology companies, including former Google engineers.

At 17, Chen could become one of its youngest team members.

Chen has not responded to requests for comment, while Discovery Loop said it does not comment on personnel matters.

The reported move therefore remains unconfirmed.

Chinese AI prodigy's departure sparks speculation

Regarded as one of China's most promising young AI talents, Chen sparked widespread speculation about his next career move after announcing last week that he would leave China for the United States.

Although he spent less than a year at Moonshot AI, Chen contributed to technical breakthroughs that helped reduce the operating costs of the company's flagship Kimi AI models.

These improvements strengthened Kimi's ability to compete with leading US AI systems, including OpenAI's GPT and Anthropic's Claude.

Moonshot AI is currently considered one of China's leading AI model developers.

Its K3 model has attracted international attention for its capabilities and open-weight architecture, which allows developers to access and use the model's parameters.

Following Chen's departure announcement, Moonshot senior researcher Du Yulun wrote on social media that he and Chen had "discussed this decision thoroughly on multiple occasions."

However, Du said the best course for Chen at his age was to have the freedom to explore and pursue new interests and intellectual curiosity as much as possible.

Discovery Loop reportedly seeks $50 billion valuation

Despite having yet to release a product, Discovery Loop is reportedly targeting a valuation of approximately $50 billion in a new fundraising round, according to a Business Insider report published last month.

The company's initial funding round was led by Radical Ventures and Khosla Ventures, with participation from Alphabet, Google's parent company.

The substantial valuation target reflects investor interest in AI systems capable of accelerating scientific discovery, although the company's commercial technology has yet to be publicly demonstrated.

At the Agentic AI Summit at the University of California, Berkeley, in early August, Jeff Dean made his first public appearance since leaving Google.

During the event, he explained that Discovery Loop's research direction was closely aligned with the concept of recursive self-improvement (RSI), a major objective pursued by leading AI companies in both the United States and China.

RSI refers to the ability of AI systems to improve their own capabilities through repeated cycles of development and refinement.

Similarly, Dean described a future in which AI could independently conduct thousands of scientific experiments in parallel, analyze the resulting data and determine which experiments should be performed next.

Such systems could potentially compress scientific research processes that traditionally take years into hours or even minutes.

"By building models that can deeply understand many different scientific and engineering disciplines, you can create a system with PhD-level expertise across multiple fields," Dean said.

The approach represents Discovery Loop's ambition to move beyond conventional AI assistants and develop systems capable of autonomously advancing scientific and technological research.

Du Lam