Intellectual Property Protection Acts as a 'Mainspring' for Artificial Intelligence

Artificial intelligence (AI) is a strategic technology leading a new round of technological revolution and industrial transformation. It represents a new focal point for international competition and a new engine for economic development. In recent years, AI companies in Hangzhou, Zhejiang Province have garnered widespread attention both domestically and internationally by continuously iterating their technologies and innovative products.

Luo Weihong, a deputy to the National People's Congress and Vice Chairperson of the Standing Committee of the Hangzhou Municipal People's Congress, has conducted in-depth research closely centered on the AI industry during her tenure. She listens to voices from the frontlines of AI development, focusing on new challenges and proposing new recommendations.

"Promoting high-quality AI development hinges on solidifying the computing power foundation and fostering synergy between algorithmic models. This is not only an inherent requirement of technological evolution but also an urgent need to build an independent, controllable, safe, and efficient modern industrial system," Luo Weihong stated. She discovered that challenges currently exist in the interplay between AI algorithms and models, including: insufficient autonomous supply capacity for core hardware, leading to high security risks in the industrial and supply chains; and the performance and efficiency of the autonomous computing power system needing improvement, making it difficult to fully meet the demands of industrial innovation and development.

During this year's National People's Congress, Luo Weihong proposed solutions for systematically promoting the coordinated development of AI algorithms and models, aiming to build an independent and controllable industrial system.

Luo Weihong outlined several key points. First, it is necessary to strengthen national strategic planning and top-level design, systematically deploy efforts to overcome core technological bottlenecks, focus on fundamental software and hardware shortcomings, implement a full-chain technological assault, enhance national coordination under a "unified national approach," and establish a two-way driving mechanism of "demand pulling and technology supply." Second, an independent and controllable high-performance computing power supply system should be built, improving infrastructure efficiency and user-friendliness by encouraging improvements in computing power efficiency, establishing national-level public service platforms for "chip-model adaptation," and promoting the development of autonomous computing power standards and software ecosystems. Finally, the collaborative innovation mechanism of "chip-model integration" needs to be deepened to create an open and win-win industrial ecosystem. This includes, for example, planning and constructing several national-level "chip-model collaborative innovation centers"; improving the financing support system for technology enterprises, particularly intellectual property pledge financing, to smooth the transformation pathway from laboratory achievements to pilot-scale amplification and then to mass production; and implementing a dual-wheel drive strategy of "open source opening" and "standards leadership."

Computing power acts as the "powerhouse" of AI, while models serve as its "smart brain." The entire innovation chain, deeply coupling the two, relies on robust intellectual property protection for key achievements and critical links.

Regarding strengthening IP protection in the AI field, the China National Intellectual Property Administration has responded to hot-button issues of widespread public concern in AI by revising the Patent Examination Guidelines and issuing the Guidelines for AI-Related Invention Patent Applications (Trial). In the process of promoting the high-quality, linked development of AI algorithms and models, what role does intellectual property play, and how should IP protection be strengthened?

In response, Luo Weihong stated: "The essence of linking AI algorithms and models lies in the collaborative innovation of 'technology + data + computing power.' In this process, intellectual property protection is extremely important and serves as the cornerstone for its healthy development. With the rapid advancement of AI, IP protection also faces new issues and challenges. Only by comprehensively strengthening IP protection in areas such as source compliance, rights layout, process control, rights relief, and institutional construction, can we protect innovation while achieving open sharing and an ecological win-win situation, thereby promoting the high-quality development of the AI industry."