_________

Leveraging its strong manufacturing base for construction machinery, the city of Changsha is reinventing itself as a hotspot of pyshical AI. No better place to train models or humanoids for the practical demands of production than the factory floor itself, city officials say.

Buildig jumanoid robots is part of this strategy, but Changsha is not competing with Shanghai oder Shenzhen for the top spot in robot dancing competitions. Rather, Zoomlion is testing them on existing production lines.

Other companies in this provincial capital with five million people are betting on AI for industrial inspection or opening their manufacturing use cases for outside developers.

As cities and regions around the world are searching for a way to log on to the industrial revolution brought upon by AI, not everybody can be the developer of frontier models.

Changsha has cleverly decided to build its AI future on its advanced manufacturing base.

Which companies in Changsha are currrently at the forefront of this effort and what their concrete strategies in physical AI look like is the topic of or Must Read this week.

_________

KUKA unveiled a platform at the China International Industry Fair (CIIF) in Shanghai on October 12 that aims to integrate AI into the control of entire factories. KAI, short for KUKA Advanced Intelligence, connects machine data, digital factory models, and AI agents that plan and carry out tasks.

KUKA aims to close a gap familiar to many manufacturers by integrating data well enough for AI to contribute useful production decisions.

Geely has just launched its new “Galaxy Zhanjian 700.” The off-road vehicle is designed to handle some of the decisions drivers face off paved roads. An AI-powered system is said to recognize the terrain and automatically switch to the appropriate driving mode.

Automatic mode selection is intended to give less experienced users an off-road experience too.

Synopsys is exploring partnerships with Chinese AI labs to connect its chip design software with local language models. AI could eventually take over tasks currently performed by conventional chip design software.

By integrating AI into its own tools, the company aims to build an additional line of business.

At Kunshan First People’s Hospital in Jiangsu province, five AI-powered robots are transporting used surgical instruments from operating rooms for reprocessing.

The robots are intended to take over the transport of contaminated instruments from hospital staff and minimize their direct contact with hazardous material.

For details and more analysis, please click here to read the full, free version of this week’s Highlights:

_________

_________

The total value of China’s core AI industry has now exceeded more than 1.2 trillion yuan (approximately 160 billion euros). More than 6,200 companies operate in the sector.

Source: News agency Xinhua, October 10, 2026.

More than 80 percent of the industrial companies in China surveyed by Siemens see opportunities for AI agents to improve efficiency and lower costs. Around 43% do not yet use them, however.

Sources: Business news service Yicai Global, October 13, 2026.

Human involvement in medical decisions could fall from nearly 100% today to less than 10% in the future, according to Duan Tao, head of the medical AI innovation center at Shanghai East Hospital.

Source: : News magazine Zhongguo Xinwen Zhoukan, October 7, 2026.

_________

China’s leadership wants to embed artificial intelligence more deeply in traditional industries. The Communist Party’s Central Committee and the State Council have issued new guidelines for developing “new quality productive forces.” They call for full implementation of the “AI Plus” initiative and the use of AI to modernize traditional industries, financial news service Cailianshe reported on October 11, 2026.

The policy term “new quality productive forces” is an umbrella term for all the emerging technologies Beijing wants to mobilize as new drivers of economic growth.

Specifically, the guidelines call for national pilot projects for industry-specific AI applications. The applications tested are to reflect local conditions and the needs of each industry.

The Chinese government is promoting new technologies such as AI, biotechnology, and robotics while also seeking to use AI to modernize its existing manufacturing base and other industries. Expected productivity gains will intensify competition from Chinese manufacturers in the years ahead.

Shanghai unveiled an action plan for “next-generation intelligent manufacturing” on October 12. By 2028, the city aims to build 500 advanced smart factories and develop more than 2,000 AI applications for manufacturing. News agency Xinhua reported the targets at the opening of the China International Industry Fair .

The targets include ten capable large language models suited to practical use and 300 AI agents. The city also plans to build infrastructure that meets the requirements of the new manufacturing processes. Industrial robot density is to rise to 600 robots per 10,000 employees.

The numerical targets may seem a little unusual to Western readers, but they are a well-established tool of Chinese industrial policy. With clear targets of this kind, the central leadership in Beijing signals to provincial and local governments that it takes AI seriously and that officials’ promotion prospects depend on their determination to meet such goals.

China’s Ministry of Industry and Information Technology and nine other agencies published a development plan for the pharmaceutical industry for the 15th Five-Year Plan period on September 18. It calls for greater use of AI in research and production, financial portal Sina reported.

Pharmaceutical companies are encouraged to work with medical institutions, research institutes, and drug research service providers to build “high-quality industry datasets and trusted data spaces for healthcare.”

The government’s main aim is to facilitate the use and exchange of pharmaceutical and healthcare data. This would provide a foundation for faster development of large AI models and agents for healthcare.

The plan covers everything from drug discovery and clinical research to production process optimization, quality management, drug distribution and use, and intelligent oversight. AI is intended to transform every stage of the pharmaceutical value chain.

The plan makes building a solid data foundation a joint task for industry, research, and medicine. Pharmaceutical companies and their software suppliers should therefore identify suitable partners for this task.

China’s National Medical Products Administration (NMPA) has published a revised draft of stricter guidelines for approving AI products and related services in China’s healthcare sector. Comments were due by October 12, 2026, according to the official website of the NMPA’s Yangtze River Delta Center for Medical Device Evaluation and Inspection.

The new draft covers both large and domain-specific AI models for medical use. It distinguishes between the use of general-purpose large models, the development of medical agents, and internally developed large models. Applicants would be expected to document risks such as hallucinations, meaning plausible-sounding false statements, and systematic bias.

New sections address multimodal products that process different types of data, as well as the generation of medical images from other imaging modalities.

Medical AI providers in China must prepare for tighter oversight. Among other requirements, the regulator is calling for risks to be documented in a way that allows them to be traced and reviewed.

_________

More than 2,000 years ago, engineers in ancient China built canals and early precursors of ship locks. Now their decendants are learning how to optimize canal traffic with artificial intelligence.

On China's new Pinglu Canal a digital twin and vertical AI models help dispatchers to coordinate shipping. And preparing a berthing plan at Qinzhou port's Honggang Terminal used to take more than an hour. Now a scheduling agent generates it within minutes.

The just-opened Pinglu Canal is one of China's most important infrastructure projects for decades. Opening a new western trade route for China's exports via ports in the Gulf of Tonkin, it will alter the flow of goods between China, ASEAN and Europe.

Improved traffic management with AI along this busy waterway offers lessons for dispatchers with scheduling tasks anywhere.

In our Deep Dive this week, we take a look at how a complex control center challenge has turned into one of industrial AI's most fascinating use cases.

We examine which AI applications help synchronize inland shipping and oceangoing vessels, how humans remain in the loop to exercise important oversight, and how even the fish in the canal are monitored in real time for their protection.