LawGenesis, a Chinese startup specializing in AI-driven cognitive security, has secured around $10 million in an angel funding round, according to people familiar with the matter.
Zhongnan Capital, Kaifeng Capital, and Planck Capital participated in the round, which will be used to develop the company’s core product, the “LawGenesis Principle,” expand its market footprint, and build a cognitive security community.
The startup is positioning itself at the forefront of a rapidly growing market. China’s AI content security sector reached roughly $2.8 billion in 2024, growing more than 35% annually, according to industry data. Traditional manual review systems are increasingly unable to keep pace with AI-generated content, which can now surge from tens to thousands of items per second.
“We’re not just replacing human reviewers,” said CEO Tan Yilang. “We’re turning them from reactive ‘firefighters’ into proactive ‘strategists.’ Our goal is to shift AI security from passive response to active defense.”
Since March 2024, LawGenesis has focused on building a continuously evolving cognitive security intelligent agent. Its flagship product, the LawGenesis Principle, reached 80% of human-level accuracy in September 2024, surpassed 95% by December, and hit 99.5% with the release of version 2.0 in March 2025. The company says it is the first in China to achieve high-precision, fully automated AI content review.
LawGenesis has partnerships with People.cn, Wuhan University’s School of Cyberspace Security, and the Wuhan Artificial Intelligence Research Institute. Its clients include platforms such as Douyin, Xiaohongshu, and Kusa AI. In one instance, deployment of the LawGenesis Principle allowed an e-commerce platform to triple efficiency in managing emerging content risks, cutting strategy adjustment time from three days to one.
The startup’s dual-engine product matrix serves both user-generated content (UGC) and AI-generated content (AIGC) markets. Its system reviews multimodal content, targets borderline cases such as financial scams, and reduces false positives, while overseas clients benefit from risk mitigation of more than 70% in text-to-image AI outputs.
“We can review complex content in 30 seconds, cutting costs by 60% compared with manual review,” Tan said. The platform supports one-click identification of 30 risk categories and can flag subtle manipulations that traditional models miss.
LawGenesis is also expanding multilingual capabilities to serve Belt and Road countries and exploring applications in news proofreading, government content screening, and consumer-level cognitive security. Its current business prioritizes enterprise clients, who account for roughly 70% of revenue, while testing consumer demand for long-term growth.
“The vision is a continuously evolving cognitive security ecosystem centered on human needs,” Tan said. “We aim to grow consumer adoption while maintaining enterprise services.”
