For three consecutive years, China's government work report has highlighted the "AI+" initiative. This year's report introduced the concept of "new forms of smart economy" for the first time.
China's AI development is moving beyond technological research and early-stage applications into a new phase of deep industrial integration and economy-wide empowerment. This move is evident in the shift from promoting "AI+" to outlining a broader smart-economy framework.
AI agents enable "one-person armies"
At the start of 2026, an emerging manifestation of this transformation is the rise of the one-person company (OPC) model. AI agents such as OpenClaw, are empowering a growing number of entrepreneurs capable of independently building commercial applications.
"Starting a business by 'raising lobsters' has become quite popular recently, and the OPC model is gaining strong momentum," said Zhang Yunquan, a member of the National Committee of the Chinese People's Political Consultative Conference (CPPCC) and a researcher at the Institute of Computing Technology of the Chinese Academy of Sciences.
The phrase "raising lobsters" has become a buzzword in the AI community, on the back of OpenClaw adopting a red lobster logo and users jokingly using the phrase to describe the process of training and developing their AI assistants.
"The emergence of AI agents such as OpenClaw and MeDo allows ordinary people without coding backgrounds to develop practical applications within a short period of time," said Ding Hong, a CPPCC member, academician of the Chinese Academy of Sciences and chair professor at Shanghai Jiao Tong University.
According to Ding, the OPC model represents a new form of "one-person army" that has become a dynamic unit within the emerging smart economy. "It is not only an agile leap by individuals in the technological wave, but also a reflection of national strategy and social transformation at the micro level," he said.
Ding highlighted the dramatic productivity gains brought about by AI agents. Tasks that previously required a team of ten people working for a week can now be completed in days — or even hours — with AI systems operating 24/7. Furthermore, AI-based products can be optimized in real time based on user feedback, improving efficiency while lowering costs.
"The OPC is not a short-term trend," Ding said. "It reflects a long-term shift in how economic activities and the social division of labor are organized."
Lower barriers empower entrepreneurs
The potential of OPCs has also attracted attention from policymakers and business leaders.
Ling Junjie, a CPPCC member and founding chairman of the Chinese Young Entrepreneurs Association in Hong Kong, believes that a number of OPC unicorn companies could emerge within the next five years.
He noted that AI tools have significantly lowered the barriers to entrepreneurship while reducing the cost of experimentation. This environment allows individuals to test ideas quickly and iterate products with far fewer resources than before.
Their defining features — light assets, rapid iteration and deep vertical focus — align well with the digital economy's operational logic, Ling said.
He encourages young entrepreneurs to actively explore sustainable business models within the rapidly expanding AI application landscape. With AI tools, ideas that previously required substantial investment can now be developed and launched more efficiently.
Policy support and challenges ahead
As the OPC trend gathers momentum, local governments across China have begun introducing policies to support the model.
Beijing recently launched an AI OPC service initiative aimed at building an incubation system through measures such as rent subsidies and computing-power support. Shenzhen plans to establish more than ten OPC communities and nurture over 1,000 high-growth AI startups by 2027. Other cities, like Suzhou and Hangzhou, are also developing OPC-focused entrepreneurial ecosystems.
Despite the enthusiasm, experts say the model still faces several challenges.
Ling pointed out legal uncertainties surrounding OPCs, including blurred boundaries between personal and corporate assets and unclear intellectual property ownership for AI-generated content.
Zhang, meanwhile, highlighted another issue: the entrepreneurial ecosystem for OPC founders can be isolated, as many individuals operate largely independently without the collaborative support systems found in traditional startups.
To tackle these issues, Ling suggested that policymakers adopt a coordinated set of measures. These could include compliance guidelines clarifying responsibilities related to AI ethics, data use and intellectual property rights, as well as the tiered access to high-quality anonymized datasets for qualified OPC developers.