In an industrial park in Zhangjiagang, a small metropolis on China’s east coast, a big buzzing and hissing machine feeds on piles of used garments and types them.
The novelty? It makes use of synthetic intelligence to kind them by composition at excessive velocity, providing a glimpse into how AI might play a job in lowering the impression of artificial textile waste.
The Fastsort-Textile machine, named one in every of Time journal’s Finest Innovations of 2025, was created by DataBeyond, a Chinese language AI recycling firm based in 2018.
“We can make full use of textile waste and reduce the amount that is incinerated which will be a great help to recycling resources,” DataBeyond CEO Mo Zhuoya mentioned.
Artificial textiles are derived from fossil fuels and are a low-cost, fashionable choice for trend manufacturing. Altogether they account for round 70% of world textile manufacturing, in response to a report from Amsterdam-based nonprofit Circle Financial system, which analyzes methods to scale back textile waste.
Textile waste is a significant world pollutant, with China because the main contributor. China led world textile exports at $142 billion, greater than double that of the European Union, in response to the World Commerce Group’s 2025 Key Insights and Developments report.
Fastsort-Textile is getting used solely in a single location in China: Shanhesheng Environmental Expertise Ltd., a textile recycling facility in Zhangjiagang that put in the machine in 2025.
The gear makes use of an AI scanner to learn the composition of such textiles and types them by fibers, after which they are often recycled.
Fastsort-Textile types by 100 kilograms (220 kilos) of garments in two to 3 minutes , in comparison with round 4 hours for one employee to do the identical factor. The machine can course of two tons per hour, whereas two individuals would want two days and at diminished accuracy, in response to evaluation by Shanhesheng.
The AI scanner measuring 5-by-2 meters (16-by-6.5 toes) works with a collection of conveyer belts. Staff load stacks of textiles onto belts that transfer them by the scanner, which emits a pointy hiss whereas studying the textiles’ composition. A stay video feed shows the studying on the scanner’s aspect.
It takes lower than one second to precisely learn one merchandise’s materials composition, which is about in response to clients’ desired benchmarks.
After the scanning course of, the textiles are transported to nylon and polyester sorting areas for recycling. Gadgets beneath the benchmark are sorted into a distinct space primarily for incineration or landfill, which is the place textile air pollution wreaks its most injury.
“This sort of thing saves money on labor costs, it saves time. When people sort materials, they can’t tell accurately if it’s 80 or 90% polyester. This machine rarely makes mistakes,” Shanhesheng Gross sales Supervisor Cui Peng mentioned.
Beforehand, as much as 50% of the processed textiles have been deemed unrecyclable and despatched to landfills or incinerated. That quantity is all the way down to 30% with the Fastsort-Textile machine, Gross sales Director Li Bin mentioned.
“Now, though machines are already capable of sorting, people’s energy is limited,” he mentioned. “People can’t work for 24 hours straight, so robots may take over the roles in the end. The ultimate goal is a ‘dark factory’ with the robots running 24 hours.”
