Lei Jun says Xiaomi must “actively embrace” the AI era as security fears swirl around OpenClaw agents

Xiaomi doubles down on AI

Lei Jun (雷军), founder and chairman of Xiaomi (小米), has publicly urged the company and the broader industry to “actively embrace the AI era” even as security researchers raise alarms about new autonomous AI agents. It has been reported that Xiaomi is moving quickly to integrate more agent-like capabilities into its devices — a strategy Lei frames as essential to keeping the company competitive as intelligent assistants become central to consumer hardware.

Security warnings from 360 Group

The push comes amid stark warnings from Zhou Hongyi (周鸿祎), founder of 360 Group (360集团), who has flagged OpenClaw — popularly nicknamed “Longxia” (龙虾) — as a growing security threat. Zhou warned that OpenClaw is not a conventional chatbot but an open‑source AI agent that can execute tasks on users’ machines: opening web pages, reading files, modifying code and orchestrating workflows. It has been reported that nearly 150,000 OpenClaw‑related assets have been detected globally and that more than 40% are located in China, a statistic experts say raises the stakes for endpoint security. Zhou urged the industry to “use AI to fight AI” and to build security as a system‑native capability so intelligent agents’ actions remain auditable and controllable.

Innovation versus risk

Lei Jun’s response underscores the tension Chinese tech firms face: accelerate AI integration or cede ground to rivals. He reportedly acknowledged the safety debate but argued that stopping innovation is not an option — instead, vendors must pair aggressive product development with stronger, baked‑in defenses. Who sets the guardrails? For Chinese companies this question comes against a backdrop of U.S. export controls, supply‑chain scrutiny and growing geopolitical attention to dual‑use AI technologies, factors that complicate both product design and international deployment.

The debate is now public and urgent. Can the industry scale agent functionality while preventing those same agents from becoming new attack surfaces? Xiaomi’s answer — and whether it implements the kind of system‑level safeguards Zhou advocates — will be watched closely at home and abroad.