In 2023, fast-fashion giant SHEIN was everywhere: Planes crisscrossed the globe, delivering parcels of the brand’s ultra-cheap clothing. Thousands of Suppliers It reached tens of millions of customers’ mailboxes across 150 countries. Influencers’ “#sheinhaul” videos promoted the company’s trending styles on social media. Billions of views.
At every step, data was created, collected, and analyzed. To manage all this information, the fast fashion industry has begun to embrace emerging AI technologies. Shein uses a proprietary machine learning application (essentially a pattern-recognition algorithm) to gauge customer preferences in real time, predict demand, and deliver it across its lightning-fast supply chain.
As AI accelerates the business of mass-producing affordable, fashionable clothing like never before, Shein Pressure mounts The company is also committed to becoming more sustainable: It has committed to reducing its carbon emissions by 25% by 2030 and reaching net-zero emissions no later than 2050.
But environmental activists and researchers say the company’s lightning-fast manufacturing methods and online-only business model are inherently emissions-intensive, and that its use of AI software to power these operations could increase emissions. These concerns are echoed by Shein’s 3rd Annual Sustainability ReportThe company’s carbon dioxide emissions are expected to nearly double between 2022 and 2023, according to a report released late last month.
“AI is enabling the transformation of fast fashion into an ultra-fast fashion industry, and Shein and Temu are pioneers of this,” said Sage Regnier, executive director of climate nonprofit Sustainable and Just Future. “They literally couldn’t exist without AI.” (Temu is a fast-growing e-commerce giant with a marketplace of products to rival Shein. Type, price, sales.
In the 12 years since its establishment, Shein has become known for its uniquely prolific manufacturing industry, with production reportedly reaching over one million tonnes per year. $30 billion By 2023, it will account for 10% of the company’s revenue. Estimates vary, but the new Shein design has as early as 10 days Up to 10,000 clothing items are added every day. The company offers the following items: 600,000 items They’re on sale all the time, with the average price being about $10. (Shane would not confirm or deny these reported figures.) One market analysis found that 44% of Gen Z in the U.S. owns at least One item Delivered monthly by Shein.
Its scale means the environmental impact will be enormous: According to the company’s sustainability report, Shain will emit a total of 16.7 million tonnes of carbon dioxide in 2023, more than the emissions of four coal-fired power plants. Spit it out in a yearThe company also Textile wastehigh level Microplastic pollutionThe report cited a range of issues, including poor manufacturing practices, exploitative labor practices, and the report said polyester, a synthetic fiber known for shedding microplastics into the environment, makes up 76 percent of all fabrics, but only 6 percent of those are recycled.
and Recent Research More than a year after the company promised to improve working conditions in its supply chain, it found that factory workers at Shein’s suppliers were typically working 75-hour weeks. While Shein’s sustainability report shows working conditions are improving, a third-party audit of more than 3,000 suppliers and subcontractors also showed that 71 percent received a score of C or below on the company’s A to E rating scale. Mediocre at best.
Machine learning plays a key role in Shein’s business model. Peter Perno-Dey, Shein’s head of global strategy and corporate affairs, told Business Insider in August that AI is Not central He suggested that this was not the case. During the presentation At a retail conference earlier this year.
“We use machine learning techniques to accurately forecast demand in a way that we believe is cutting edge,” he said. Pernod-Day told the audience that all of Shein’s 5,400 suppliers have access to an AI software platform that gives them up-to-date information on customer preferences, allowing them to change production accordingly in real time.
“This means we produce very small numbers of each garment,” he said. “That means we have very little waste and very little inventory wastage.” On average, the company says it stocks 100 to 200 copies of each item. That’s in contrast to traditional fast-fashion brands, which produce thousands of each item per season and try to predict trends months in advance. Shein calls its model “on-demand,” but technology analysts say Speaking to Vox in 2021, This was called “real-time” retail.
During the conference, Pernoday also suggested that the technology is helping the company pick up on “micro-trends” that customers want to wear. “We can detect that and act on it. This is where I think we’re really pioneering,” he said. Recent class action lawsuits The plaintiffs in the New York District Court allege that the company’s AI market analysis tools are being used in an “industrial-scale scheme of coordinated digital piracy against the work of small designers and artists,” scraping designs from the internet and sending them directly to factories for production.
In an emailed statement to Grist, a Shine spokesperson reiterated Peter Pernod Dey’s assertion that technology is enabling the company to reduce waste and become more efficient, and suggested that the company’s emissions increase in 2023 would be due to strong business performance. “We do not believe that growth is at odds with sustainability,” the spokesperson said.
An analysis of Shein’s sustainability reports by trade publication Business of Fashion found that last year the company’s emissions were roughly Double your profit margin This makes Shein the fashion industry’s biggest emitter. By comparison, Zara’s emissions only increased by half as much as its sales. Other industry giants, including H&M and Nike, saw their sales grow but their emissions decline from the previous year.
Shen Lu, a professor of fashion and apparel at the University of Delaware, said SHEIN’s emissions are particularly high because of its reliance on air transport. “AI has a wide range of applications in the fashion industry, so it’s not necessarily a bad thing,” Lu said. “The problem lies in the nature of SHEIN’s particular business model.”
Other big brands ship goods overseas in bulk, preferring lower-cost ocean freight and having suppliers and warehouses in numerous countries to shorten the distance goods travel to reach consumers.
According to the company’s sustainability report, 38% of Shain’s climate footprint comes from transportation between its facilities and customers, and the remaining 61% comes from other parts of its supply chain. The company is based in Singapore and has suppliers in several countries, but the majority of its clothing is made in China and sent to customers by airmail in individually addressed packages. In July, the company sent about $15,000 worth of goods to a customer. 900,000 These are being sent to the United States every day.
Mark Asensio/NurPhoto via Getty Images
A Shane spokesperson told Grist that the company Decarbonization The company has developed a roadmap to address its supply chain footprint. Recently, the company Increased inventory Storing goods in U.S. warehouses allows the company to offer faster delivery times to American customers, which has led to increased use of cargo ships. More carbon efficient than cargo planes.
“Reducing carbon emissions in the fashion industry is a really complicated process,” Lu said, adding that many brands are using AI to streamline their operations. “It really depends on how you use it.”
Research suggests that using certain AI techniques could make companies more sustainable. “That’s the missing piece,” says Sharia Akhter, associate dean of business and law at the University of Wollongong in Australia. In May, Akhter and her colleagues published a paper titled “A New Approach to AI.” the study When fast fashion suppliers used AI data management software to comply with big brands’ sustainability goals, those companies saw their profitability increase and emissions reduced. Atkar says the technology’s main use is to closely monitor environmental impacts like pollution and emissions. “This kind of tracking wasn’t available before the advent of AI-based tools,” he says.
Shain did not respond to a request for comment about whether it uses machine learning data management software to track emissions, which is one of the uses of AI included in Akhtar’s study, but the company’s highly touted use of machine learning software to forecast demand and reduce waste is another.
Either way, the company has a long way to go before it hits its goal: Grist calculated that Shain’s reported emissions savings in 2023, thanks to measures like providing solar panels to suppliers and opting for ocean shipping, would amount to about 3% of the company’s total annual carbon emissions.
Sustainable and Just Future’s Regnier believes that ethical use of AI is impossible in the fast fashion industry. He said the largely unregulated technology allows brands to intensify harmful impacts on workers and the environment. “People who work in fast fashion factories are now under incredible pressure to produce more, faster,” he said.
Regnier and Lu believe the key to a more sustainable fashion industry is to convince customers to buy less. Lu said that if companies use AI to boost sales without changing their unsustainable practices, their climate impact will be correspondingly larger. “The overall effect of this is that you can offer more popular products on the market and encourage consumers to buy more than they did before,” he said. “Of course, the overall carbon impact will be larger.”
