Guozhen AIGlobal AI field notes and model intelligence

Realtime AI News

StepFun enters AI agent race, Chairman says "future OS must be cross-device"

StepFun (Jieyue Xingchen) chairman Yin Qi announced the company's entry into the AI agent arena, stating that the future operating system "must be cross-device." This signals the Chinese AI company's strategic expansion from large language model development toward an agent-oriented operating system layer.

Published
阶跃星辰入局智能体赛道,印奇称"未来的OS一定是跨端的"
Image source: huggingface.co

According to a report from QbitAI, StepFun (Jieyue Xingchen) chairman Yin Qi stated at a recent public event that the company is entering the AI agent space and proposed a vision for the future operating system — "the future OS must be cross-device." This strategic declaration signals that StepFun is extending from large model development toward a higher-level agent platform.

StepFun is one of the most closely watched emerging AI companies in China, previously known primarily for its self-developed large language models. Yin Qi, a co-founder of computer vision company Megvii, brings deep technical and industry expertise. His remarks about an agent OS suggest StepFun may be building an AI operating system layer spanning multiple endpoints including phones, PCs, and vehicles.

The "cross-device" concept is noteworthy. Unlike traditional single-device operating systems, a cross-device OS would allow AI capabilities to flow seamlessly across different hardware — a task started on one device could be continued on another, with the agent handling coordination and orchestration.

This direction aligns with global AI industry trends. From OpenAI's Agent plans to Anthropic's Computer Use feature to similar moves by domestic Chinese players, agents are widely seen as AI's next critical battleground. Competition at the operating system level determines who dominates the underlying ecosystem.

Yin Qi did not reveal a specific timeline or product details for StepFun's agent OS. However, given the company's technical foundation, industry observers speculate it may build a cross-device AI Agent framework on top of its proprietary large models, focusing primarily on mobile and in-vehicle scenarios.

StepFun's entry intensifies competition in China's agent track. From model capabilities to platform ecosystems, from single products to OS-level layouts, AI companies are racing to find differentiating paths.

Why it matters

StepFun's entry into the agent OS race intensifies competition in China's AI platform layer, with cross-device capability emerging as a potential differentiator for future agent products.

阶跃星辰StepFun智能体操作系统
Back to realtime news

Nearby Updates

All

07/15, 02:33

Google faces class-action lawsuit from major publishers over Gemini AI training on copyrighted works

Hachette, Cengage, Elsevier, and other publishers have filed a class-action lawsuit against Google, alleging the company trained its Gemini AI on copyrighted works without permission and intentionally removed copyright information to conceal the infringement. The case was filed in the Southern District of New York.

07/15, 01:49

Google localises Gemini AI for regulated sectors in India, expands enterprise push with offline deployment

Google announced Indian enterprises and public sector organizations can now run Gemini AI models within Indian data centers via Google Distributed Cloud, with fully offline deployment supporting data localization requirements. At I/O Connect India 2026 in Bengaluru, Google also unveiled security tools for agentic AI including the Sec-Gemini V3 cybersecurity agent and the open-source CAPSEM secure runtime environment.

07/15, 03:39

Hinge founder raises $18M for AI-powered audio-first dating service Overtone

Justin McLeod, the founder of Hinge, has raised $18 million for his new AI dating company Overtone. The service describes itself as a voice- and audio-forward platform that uses AI to provide highly curated introductions, launching later this year in select locations.

07/15, 03:39

Claude is significantly more polite in Hindi and Arabic, researchers find

Researchers have discovered that Anthropic's Claude AI assistant shows markedly different behavior across languages, responding more politely and warmly in Hindi and Arabic compared to English. The finding raises important questions about consistency of AI safety alignment across different languages.