Key Highlights
- Austrian developer Peter Steinberger’s OpenClaw AI agent has triggered a viral phenomenon in China, with adoption rates now exceeding those in the United States.
- Major technology companies including Baidu and Tencent are organizing large-scale public demonstrations to assist ordinary citizens in configuring the software.
- Chinese adopters have dubbed the technology “raising a lobster” and are leveraging it for solo entrepreneurship ventures, stock selection, and workflow automation.
- Regional authorities are providing financial incentives reaching 20 million yuan (approximately $2.8 million) annually for approved “one-person company” initiatives.
- Government regulators alongside financial institutions, academic institutions, and state agencies have issued advisories against deployment due to data privacy vulnerabilities.
The artificial intelligence landscape in China reached unprecedented excitement levels this year following the emergence of OpenClaw, an open-source autonomous AI agent developed by Austrian programmer Peter Steinberger. This sophisticated software possesses capabilities to operate computers independently, navigate internet resources, purchase airline reservations, and coordinate multiple bot systems — all functioning without direct human intervention.
Nvidia Chief Executive Jensen Huang described it as “the next ChatGPT.” Within China’s borders, the technology has evolved into a nationwide phenomenon.
Chinese users affectionately refer to the application as “lobster,” transforming its installation process into a communal experience. Technology corporations such as Baidu and Tencent have organized mass public gatherings where hundreds of attendees queue for assistance installing the application on their mobile devices and computers.
“Everyone in my circle — colleagues and acquaintances — seems to have adopted it,” explained Gong Sheng, a recent adopter who participated in a Baidu demonstration in Beijing. “I refuse to fall behind the curve.”
Following its initial release in November 2025, OpenClaw achieved recognition as among the most rapidly expanding initiatives throughout GitHub’s existence, the dominant platform for software developers globally.
Cybersecurity analysts at US-based SecurityScorecard discovered that Chinese adoption of OpenClaw has already overtaken American usage rates.
Practical Applications Driving Adoption
Chinese adopters have discovered diverse applications for this technology. Many are establishing what local media terms “one-person companies” — compact commercial operations functioning almost exclusively through artificial intelligence.
“Traditional employees require downtime and rest, but OpenClaw operates continuously around the clock,” explained Wang Xiaoyan, an entrepreneur utilizing the agent for her commercial venture.
Additional users deploy the system for equity selection, lottery number purchasing, electronic commerce store creation, and revenue-generating application development.
Municipal governments are actively promoting these initiatives. Several jurisdictions now offer financial support packages worth up to 20 million yuan annually for approved solo-operator businesses constructed using artificial intelligence platforms.
Retired professionals and university students have flocked to configuration workshops, seeking supplementary income opportunities. During a training session conducted by AI company Zhipu in Beijing, 60-year-old Fan Xinquan described training an agent to catalog his professional expertise more effectively than conversational platforms like DeepSeek.
This movement corresponds with China’s comprehensive AI Plus national strategy, designed to integrate artificial intelligence throughout every economic sector.
Regulatory Concerns and Escalating Expenses
Universal enthusiasm remains elusive. Chinese regulatory bodies have intensified cautions regarding information security and privacy vulnerabilities associated with OpenClaw.
China to restrict OpenClaw use at state-owned firms, government agencies, and some major banks over security concerns, according to Bloomberg.
Some employees were told not to install the AI agent on office devices or personal phones connected to work networks. pic.twitter.com/Jn3zYUQQ4g
— Wall St Engine (@wallstengine) March 11, 2026
Government departments, banking institutions, securities firms, and educational establishments have prohibited staff members from deploying the software. China’s state-controlled People’s Daily released an editorial calling upon authorities to “resolutely uphold the security baseline.”
End users have voiced apprehensions as well. “Ordinary individuals like ourselves struggle to comprehend what permissions we’ve granted and what information it has accessed,” noted user Gong Zheng.
Practical challenges have also surfaced. Artificial intelligence startup Zhipu increased token pricing for its OpenClaw-compatible model by 20% during the current week.
A widely circulated post on Chinese social platform Rednote, headlined “Goodbye OpenClaw,” documented how everyday users invested substantial funds purchasing tokens only to receive “an accumulation of worthless information.”
At a recent Baidu demonstration event, a live presentation featured an OpenClaw agent processing a voice instruction to purchase coffee through a McDonald’s application via connected smart technology. The transaction required nearly two minutes to complete — underscoring the disparity between the platform’s theoretical capabilities and its present-day operational performance.
