In an industrial park in Zhangjiagang, China, a large machine feeds on piles of used clothing and sorts them by fiber composition. The system, Fastsort-Textile, uses artificial intelligence to separate garments based on what they are made of—part of an effort to reduce the amount of textile waste that ends up incinerated or landfilled.
Fastsort-Textile was created by DataBeyond, a Chinese AI recycling company founded in 2018, and was named by Time magazine as one of its Best Inventions of 2025. DataBeyond CEO Mo Zhuoya said, “We can make full use of textile waste and reduce the amount that is incinerated which will be a great help to recycling resources,” describing the machine’s purpose as improving how recycling materials are recovered.
Synthetic textiles derived from fossil fuels make up a large share of global fashion production, the AP reported, citing an Amsterdam-based nonprofit, Circle Economy. The AP also reported that China is a major contributor to textile waste, and said China led global textile exports at $142 billion—more than double the European Union’s level—according to a 2025 World Trade Organization report.
The Fastsort-Textile system is installed only at one site in China so far: a textile recycling facility operated by Shanhesheng Environmental Technology Ltd. in Zhangjiagang, where the company installed the equipment in 2025. Shanhesheng uses an AI scanner that reads a textile’s composition and then routes the items to the appropriate recycling stream.
Shanhesheng said the machine can sort through 100 kilograms (220 pounds) of clothes in two to three minutes, compared with around four hours for one worker to do the same job. The company also said the equipment can process two tons per hour, while two people would need two days and would work at reduced accuracy, according to the analysis Shanhesheng provided to the AP.
The scanner itself measures 5 by 2 meters (16 by 6.5 feet) and works with conveyer belts to move stacks of textiles through it, according to the report. As the material passes through, the scanner emits a sharp hiss while reading the clothing’s composition, and the system shows a live video feed on the scanner’s side. Shanhesheng said it takes less than one second to accurately read the material composition of an item, and that the composition reading is set according to customer benchmarks.
After scanning, textiles move to nylon and polyester sorting areas for recycling, while items below the benchmark are sent to a different area, mainly for incineration or landfill. Shanhesheng Sales Manager Cui Peng said in an interview, “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,” pointing to differences between machine readings and manual judgment.
Shanhesheng also said prior to using the machine, up to 50% of processed textiles were deemed unrecyclable and were sent to landfills or incinerated. With Fastsort-Textile in operation, the company said the figure dropped to 30%, with Sales Director Li Bin saying, “Now, though machines are already capable of sorting, people’s energy is limited.” He added, “People can’t work for 24 hours straight, so robots may take over the roles in the end,” and described a longer-term goal of “a ‘dark factory’ with the robots running 24 hours.”