How AI is transforming recycled packaging

That shift is driving rapid growth in advanced digital technologies across the packaging value chain. Machine learning, computer vision, digital twins and predictive analytics are helping companies improve recycling performance and better understand how packaging behaves in real-world systems. 

More importantly, these technologies are helping businesses understand the difference between packaging that is technically recyclable and packaging that is actually recycled in practice. 

Why AI in recycled packaging is the new trend 

Governments worldwide are introducing legislation targeting plastic reduction, recycled content targets and extended producer responsibility (EPR). At the same time, consumers increasingly expect brands to prove sustainability claims with measurable action. Major packaging and consumer goods companies are also setting ambitious circularity goals focused on reducing virgin plastic use and increasing recycled material recovery. 

Artificial intelligence is becoming a key enabler for this transition. Advanced analytics and machine learning can process enormous amounts of packaging and recycling data, helping companies improve sorting accuracy, reduce contamination, optimise packaging design and predict recycling performance before products even enter the market. 

As a result, packaging decisions are increasingly being guided by real-world data rather than assumptions about recyclability. 

Top applications of AI in packaging today 

Advanced digital technologies are helping businesses move beyond theoretical recyclability towards measurable packaging performance throughout the entire lifecycle, from production to recovery and reuse. 

1. Waste sorting 

Traditional recycling systems often struggle with contamination and complex packaging formats. New sorting systems combine computer vision, optical sensors and deep learning models to identify materials based on shape, colour, transparency, texture and packaging design.  

Rather than relying solely on markers or barcodes, these systems can recognise plastics, paper, food-grade materials and contaminants at high speed, improving recovery rates and reducing landfill waste. 

The UK is emerging as a leader in AI-enabled recycling infrastructure. In Coventry, the new Sherbourne Materials Recycling Facility (MRF) uses artificial intelligence, robotics and optical sorting technologies to process up to 175,000 tonnes of household recycling annually. Jointly owned by eight local authorities, the facility can separate paper, plastics, glass and metals to purity rates of up to 99%, enabling more material to be recycled within the UK rather than exported overseas. The centre demonstrates how digital technologies are transforming recycling from a labour-intensive process into a highly automated, data-driven system. 

Similar technologies are being deployed globally. Companies such as AMP Robotics deploy robotic systems to identify and sort recyclable materials in waste facilities at high speeds, while TOMRA Recycling’s GAINnext deep-learning technology improves the sorting of food-grade and non-food-grade plastics, helping recover higher-quality recycled materials suitable for reuse in packaging applications. 

TOMRA has also introduced a waste analytics platform that interprets recycling streams, identifies inefficiencies and provides actionable insights to improve recovery performance and end-to-end material tracking. 

2. Material innovation 

Machine learning models can analyse thousands of material combinations and formulation parameters to identify recyclable, compostable and bio-based alternatives far faster than traditional trial-and-error research methods. 

Companies including Nestlé and IBM have explored generative design tools to identify alternative packaging materials that maintain product protection while improving recyclability and reducing environmental impact. 

3. Packaging design 

Digital twins and simulation tools are helping companies design packaging with recycling performance in mind from the earliest stages of development. 

Brands can now simulate how packaging performs during manufacturing, transportation, consumer use and recycling. This allows companies to reduce unnecessary material use while improving recyclability. 

Companies such as Unilever are using digital twins to test packaging before starting production, helping to evaluate how well it moves through sorting and recycling infrastructure. 

In 2026, Germany-based startup one.five raised €14 million to expand its platform, which helps packaging companies develop products with product-market fit built into the design process. 

4. Consumer demand analytics 

Predictive analytics can help businesses forecast packaging demand more accurately, optimise logistics and reduce overproduction. By improving demand planning, companies can minimise packaging waste and better align packaging choices with circular economy objectives. 

Amazon uses machine learning, computer vision and natural language processing to predict the most appropriate packaging for individual products and orders, reducing packaging use and waste. 

5. Smart packaging 

Sensors and intelligent packaging systems are opening up new opportunities for packaging innovation. These technologies can track freshness, monitor temperature and reduce food waste across supply chains.

Mimica’s smart freshness labels respond to actual product conditions rather than estimated expiry dates, helping to reduce unneccesary food waste. 

Digital watermarking is also gaining momentum. Invisible digital markers embedded into packaging can help automated sorting systems identify materials more accurately, improving recovery quality and reducing contamination without compromising branding or product appearance. For example, the HolyGrail 2.0 initiative has demonstrated that digital watermarks can enable high-precision sorting of packaging waste at industrial scale, helping create higher-quality recycling streams. 

6. Traceability 

One of the fastest-growing applications is helping brands understand what actually happens to packaging after disposal. 

Greyparrot has developed waste intelligence systems that monitor materials moving through recycling facilities using camera systems installed above conveyor belts. The technology analyses waste streams in real time, helping facilities better understand material composition, contamination and recovery rates. 

Major brands are increasingly using these insights to assess whether their packaging is performing as intended within existing recycling infrastructure. 

For example, Greyparrot partnered with Kenvue to evaluate how packaging performs in commercial-scale recycling systems, helping identify opportunities to improve packaging design and increase recovery rates. The company has also worked with Amcor, Unilever and Asahi Beverages to generate real-world recycling data that can be used to support packaging circularity goals. 

This represents a significant shift for the industry. Historically, recyclability has been assessed through lab testing or technical specifications. Waste intelligence platforms now allow companies to measure actual recycling performance, helping them understand which packaging formats are successfully recovered and where materials are being lost within the system. 

Challenges and limitations 

Despite rapid growth, several challenges remain. 

• Advanced sorting systems and robotics require substantial upfront investment, making adoption difficult for smaller recycling facilities. 

• Packaging complexity also remains a major issue. Multi-layer materials, flexible plastics and composite packaging are still difficult to process efficiently, even with increasingly sophisticated sorting technologies. 

• Another challenge is infrastructure inconsistency. Recycling systems vary widely between countries and regions, limiting the scalability of some solutions. 

• These technologies also depend heavily on high-quality data. As packaging formats evolve, algorithms must continuously adapt and retrain to maintain accuracy. 

• There are also growing discussions around the environmental footprint of large-scale computing systems and the need to balance digital innovation with sustainability goals. 

How CPI can support you 

At CPI, we enable businesses to develop innovative packaging solutions tailored for a circular economy. Whether you're exploring data-driven design tools or testing next-generation sustainable materials, our expertise, advanced facilities and strong industry partnerships can help accelerate your innovation journey.

Our core capabilities include:

• Data-driven materials and product design – using modelling and simulation tools to optimise performance, material use and sustainability outcomes 

• Data generation for model development – generating high-quality experimental datasets to train, validate and improve predictive models for materials development 

• Sustainable polymer development – accelerating the development of bio-based, recyclable and compostable materials through rapid formulation and testing 

• Performance testing and validation – ensuring packaging meets durability, barrier and shelf-life requirements 

• Lifecycle assessment and circularity analysis – understanding environmental impacts and identifying opportunities for improvement 

• Pilot-scale manufacturing facilities – bridging the gap between laboratory innovation and commercial production. 

Future outlook 

The future of AI in recycled packaging looks highly promising. 

Industry experts increasingly believe that artificial intelligence and advanced analytics will become central to circular packaging systems over the next decade. Recycling facilities are expected to become more automated, while brands will rely more heavily on real-world recycling data to guide packaging decisions. 

One of the biggest shifts already underway is the move from designing packaging purely for cost and shelf appeal towards designing packaging for recovery and circularity. 

Sustainable packaging is no longer just about making packaging technically recyclable. The real challenge is ensuring materials are recovered, sorted and reused at scale. 

The companies that gain the greatest advantage will be those that use digital technologies not only to improve recycling operations, but also to design packaging that performs effectively throughout the entire circular economy.