The global fashion industry produces a staggering amount of waste. Every year, 92 million tonnes of textile waste are generated worldwide. That is equivalent to a garbage truck full of clothing being incinerated or sent to a landfill every second. Between 2000 and 2015, clothing production doubled, while the average lifespan of garments fell by 36%. Yet only 8% of textile fibers used in 2023 were recycled.

The data shows that the recycling challenge is clear. Traditional manual sorting methods rely on human vision and touch, making them too slow and inconsistent to handle the growing volume and complexity of textile waste. Artificial intelligence, robotics, digital identification, and automated sorting technologies are helping recycling facilities process materials faster and more accurately.

The numbers reflect this shift. According to King Research, the global textile recycling market size was valued at USD 5.55 billion in 2025 and is projected to grow from USD 5.90 billion in 2026 to USD 9.40 billion by 2033, at a CAGR of 6.89% over the forecast period. This growth is driven by environmental concerns, rising volumes of textile waste, and heightened regulatory focus on sustainable waste management practices.

Governments and apparel brands are investing heavily in digital recycling infrastructure. The European Union's Extended Producer Responsibility (EPR) requirements and the introduction of Digital Product Passports are encouraging wider adoption of AI and automation across textile recycling operations.

Why Textile Recycling is Becoming Complex

Textile recycling was once relatively straightforward. Most garments were made from a single natural fiber, such as cotton or wool, making them easier to shred and mechanically recycle. Today's textile waste stream is far more complex.

The rapid growth of fast fashion has changed the composition of discarded clothing. Manufacturers now design garments primarily for affordability and performance rather than recyclability. As a result, modern textiles often incorporate multiple materials and treatments, complicating recycling.

These include:

• Blended fabrics: Cotton-polyester blends are common, but separating the different fibers remains technically challenging.
• Synthetic fiber: Polyester now accounts for approximately 70% of global textile production.
• Dyes, coatings, and chemical treatments: These materials make fiber identification and recycling more difficult.
• Non-textile components: Buttons, zippers, labels, and decorative embellishments must be removed before recycling begins.

Why is Textile Recycling Difficult?

Textile recycling is difficult because most modern garments contain blended fibers, dyes, and chemical treatments that are difficult to separate. Fast fashion has increased both the volume and complexity of textile waste. Traditional manual sorting methods cannot accurately identify fiber composition at scale, leading to contamination and lower-quality recycled materials.

The variety of textile waste is one of the industry's biggest challenges. According to industry research, much of the discarded clothing contains blended fibers, dyes, and finishing treatments that are costly and difficult to separate. Buttons, zippers, labels, and decorative embellishments add another layer of complexity because recyclers must remove these non-textile components before processing.

The scale of the challenge is equally significant. Europe generated 13.3 million tonnes of post-consumer textile waste in 2025, yet only 1.5 million tonnes were collected and sorted. That equals roughly one tonne out of every nine tonnes generated. Worldwide, only 12% of textile waste is recycled.

Why Manual Sorting is No Longer Enough

For decades, textile sorting has relied on manual labor. Workers inspect garments visually and by touch to identify fiber types before separating them into different categories. As textile waste volumes continue to rise, this approach has become increasingly inefficient.

Limitations of Manual Sorting

The productivity difference is substantial. DataBeyond's Fastsort-Textile machine, recognized as one of Time magazine's Best Inventions of 2025, sorts 100 kilograms of clothing in just two to three minutes. A worker would typically require around four hours to complete the same task. The machine processes two tonnes per hour, while two workers would need nearly two days to achieve the same output with lower accuracy.

How AI is Transforming Textile Sorting

Artificial intelligence is changing the way recycling facilities identify, classify, and process textile waste. Modern systems combine machine vision, spectroscopy, robotics, and machine learning to improve sorting accuracy and throughput.

AI-Powered Fiber Recognition

AI-powered fiber recognition uses computer vision and deep learning models to identify textile composition. Research has demonstrated more than 95% classification accuracy for pure fiber while showing encouraging results for blended materials.

Near-Infrared (NIR) Spectroscopy

Near-infrared (NIR) spectroscopy identifies fiber composition by analyzing each textile's spectral signature without damaging the material. When paired with machine learning models, particularly Convolutional Neural Networks (CNNs), NIR systems deliver highly accurate fiber classification.

Hyperspectral Imaging and Digital Twins

Hyperspectral imaging combined with digital twin technology provides another level of automation. Recent research has shown that integrating Visual Language Models (VLMs) with grasp detection and digital twin technology enables scalable, autonomous textile sorting.

The Role of Automation Beyond Sorting

AI-powered sorting is only one part of the recycling process. Automation is improving efficiency across the entire textile recycling value chain.

Material Handling

Automated bale dewiring and infeeding systems supplied by Cross Wrap for Circulose's Ortviken plant improve both safety and operational efficiency during the first stage of recycling. The Ortviken facility is recognized as the world's first commercial-scale chemical textile recycling plant.

Robotic Separation

Robots equipped with AI vision systems identify and remove buttons, zippers, logos, and other non-recyclable components. Researchers at the Rochester Institute of Technology are developing fully automated systems that dismantle used garments before recycling.

Quality Inspection

International technology group ANDRITZ has launched an automated, AI-driven inspection system that accurately rates the structural quality of post-consumer textiles before they enter the recycling loop. By evaluating fabrics at the yarn and fiber levels in under one second, the system generates a standardized metric, the "teXscore," to match textile waste with the most viable recycling methods.

Conveyor Systems and Bale Management

Companies such as PICVISA and Girbau have developed fully automated sorting lines that combine automated feeding, optical sorting, and material classification into a continuous recycling process.

Why AI is Accelerating Chemical Textile Recycling

Chemical recycling represents one of the most advanced recycling methods available today. Unlike mechanical recycling, which shortens fiber and reduces material quality, chemical recycling breaks textiles down to the molecular level. This process enables true fiber-to-fiber recycling while maintaining virgin-quality output.

AI supports chemical recycling in several ways.

• Fiber identification: AI determines the precise fiber composition before processing.
• Feedstock consistency: Automated sorting provides consistent, high-quality input materials.
• Process optimization: AI continuously monitors and adjusts recycling parameters in real time.

Technology Comparison

Industry Examples: How Companies Are Using AI and Automation

Real-world deployments show how AI and automation are improving textile recycling across different regions and recycling processes.

PICVISA and Girbau

Spanish companies PICVISA and Girbau have partnered to build a fully automated textile recycling plant in Northern Europe. The facility combines Girbau's Sortech automated feeding system with PICVISA's ECOSORT optical sorting technology. It is one of the first plants in Europe to integrate automated feeding and advanced optical sorting for post-consumer textiles. With this installation, ECOSORT technology is now operating at eight sites across Europe.

Salvos Stores (Australia)

In March 2026, Salvos Stores opened Australia's first AI-enabled textile recovery facility in Ipswich with USD 4.9 million in government funding. The automated facility uses AI and robotics to sort textiles by material type and remove buttons, zippers, and other non-textile components. It can process up to 5,000 tonnes of used clothing and textiles each year.

Avery Dennison and TEXAID

In June 2023, Avery Dennison and TEXAID partnered to pilot digital identification technologies for textile recycling. The project uses Avery Dennison's atma.io platform to assign digital identities to garments, making it easier to track products throughout the recycling process and prepare for future EU regulations.

Why Digital Product Passports Could Make AI Even More Effective

One of the most significant developments in textile recycling is the introduction of Digital Product Passports (DPPs). Beginning in 2026, the European Union will require textile products sold within its market to include a Digital Product Passport that can be accessed through a QR code, NFC chip, or similar digital identifier.

Access Strategic Insights on the Textile Recycling Market

Explore market forecasts, technology innovations, investment trends, regulatory developments, and growth opportunities driving the next generation of textile recycling.

What is a Digital Product Passport?

A Digital Product Passport is a digital record that follows a product throughout its lifecycle. It stores information about the product's material composition, manufacturing process, supply chain, repair history, and sustainability data.

Why Digital Product Passports Matter

• Material transparency: DPPs provide accurate fiber composition data, reducing guesswork during sorting.
• Improved sorting accuracy: AI systems can combine DPP data with optical scanning to improve material identification.
• Traceability: DPPs allow materials to be tracked throughout the recycling value chain.
• Regulatory compliance: They support compliance with the EU's Ecodesign for Sustainable Products Regulation (ESPR).

The Digital Product Passport forms a key part of the EU's Ecodesign for Sustainable Products Regulation and supports the broader objectives of the European Green Deal. Requirements for textile products are expected to be published in January 2026, with enforcement beginning in July 2027.

Challenges Slowing AI Adoption

Although AI offers significant benefits, several factors continue to slow adoption across the textile recycling industry.

Capital Investment

Initial investment remains one of the largest barriers. AI-powered sorting systems require significant spending on hardware, software, and system integration. However, the productivity gains can justify the investment. For example, DataBeyond's Fastsort-Textile machine processes two tonnes per hour, compared with approximately two days of work for two manual sorters.

Technology Integration

Many recycling facilities still rely on equipment that was not designed for automation. 

Data Standardization

AI systems depend on consistent product information. Without standardized identification methods and data formats, communication across the recycling chain becomes difficult. Digital Product Passports are expected to address this challenge by providing standardized product information.

What Textile Recyclers Should Consider Before Investing in Automation

Before investing in AI and automation, textile recyclers should assess several operational factors.

• Waste volumes: Does the facility process enough material to justify automation? Facilities handling fewer than 5,000 tonnes annually may struggle to achieve a positive return on investment.
• Material diversity: Facilities processing highly mixed textile waste usually benefit the most from AI-powered sorting.
• Labor costs: Automation generally delivers faster payback in regions with higher labor costs.
• Existing infrastructure: Determine whether current equipment can support automation or requires significant upgrades.
• Chemical recycling capability: Precise sorting provides the greatest value for facilities that use or plan to use chemical recycling.
• Digital readiness: Assess whether operations can integrate with Digital Product Passport systems.

Is Your Recycling Operation Ready for AI?

Consider the following questions:

• Do you process large volumes of mixed-fiber textile waste?
• Are manual sorting costs increasing?
• Does your organization have circular textile or sustainability targets?
• Are you planning to invest in chemical recycling?

The Future of AI in Textile Recycling

The next decade is expected to bring major advances in textile recycling through wider adoption of AI and automation.

• Fully automated recycling facilities: Plants similar to the PICVISA-Girbau facility are expected to become increasingly common.
• AI-powered quality control: Systems such as ANDRITZ's quality inspection technology will improve consistency and reduce contamination.
• Advanced robotics: Robots will identify and remove non-recyclable components with greater speed and precision.
• Predictive maintenance: AI will identify potential equipment failures before they interrupt operations.
• Digital twins: Virtual facility models will support process optimization, testing, and workforce training.
• Closed-loop recycling: AI will improve fiber recovery, allowing textiles to be recycled into new textiles with comparable quality.
• Smart circular manufacturing: AI will support material tracking and optimization across the entire textile lifecycle, from design through recycling.

How Will AI Transform Textile Recycling Over the Next Decade?

Over the next decade, AI will drive the development of fully automated textile recycling facilities capable of sorting mixed-fiber waste accurately at an industrial scale. These systems will direct materials to the most suitable recycling processes and produce high-quality recycled fiber. Combined with Digital Product Passports and advances in chemical recycling, AI will support true closed-loop textile recycling.

The European Union's AUTOLOOP project, launched in September 2025, highlights this direction. The initiative aims to develop a recycling system capable of processing up to 1.24 million tonnes of textile waste annually by 2050 while creating more than 130,000 green jobs across Europe. Fraunhofer UMSICHT coordinates the project with 13 partners from across the region.

Frequently Asked Questions About Textile Recycling

Why is textile recycling difficult?

Textile recycling is difficult because modern garments contain blended fibers, dyes, and chemical treatments that are difficult to separate. Fast fashion has increased both the volume and complexity of textile waste, while manual sorting cannot accurately identify fiber composition at scale.

How is AI used in textile recycling?

AI improves textile recycling by using machine vision and NIR spectroscopy to accurately identify fiber composition and assess material quality. It further enhances efficiency by automating sorting decisions and optimizing chemical recycling processes

What is automated textile sorting?

Automated textile sorting combines AI, machine vision, NIR spectroscopy, and robotics to identify and separate textiles by fiber type, color, and composition with greater speed and accuracy than manual sorting.

Can AI identify textile fiber?

Yes. AI systems that combine NIR spectroscopy with deep learning achieve more than 95% classification accuracy for pure fiber. They can also identify blended materials with high accuracy.

What is the future of AI in textile recycling?

The future includes fully automated recycling facilities, AI-powered quality control, advanced robotics, predictive maintenance, digital twins, and closed-loop recycling systems that convert used textiles into new textiles with comparable quality.

Conclusion

Textile recycling has reached a critical stage. The industry generates tonnes of textile waste each year, yet a smaller percentage of textile fiber comes from recycled sources. Traditional manual sorting methods cannot keep pace with the growing volume or increasing complexity of textile waste.

AI and automation are changing this reality. AI-powered fiber recognition can identify materials with more accuracy, while automated sorting systems can process up to two tonnes of textiles per hour. 

Chemical recycling is creating new opportunities for fiber-to-fiber recycling. Advanced AI sorting provides cleaner feedstock, improving recycling efficiency and producing virgin-quality recycled fiber. Regulation is also driving adoption. The European Union's revised Waste Framework Directive introduces mandatory Extended Producer Responsibility (EPR) schemes for textiles, while Digital Product Passports will improve material traceability and sorting accuracy throughout the recycling process.

Companies that invest in AI-powered recycling technologies today will be better prepared for future regulatory requirements and circular economy goals. As textile waste continues to grow, AI and automation will play an increasingly important role in making textile recycling more efficient, scalable, and commercially viable.

Download the textile recycling market report to explore market trends, automation investments, company strategies, regional opportunities, and future forecasts shaping the global textile recycling industry.

Speak with a circular economy industry analyst to learn how AI-powered textile recycling technologies can support your organization's sustainability goals and long-term growth strategy.