Most of the devices we use every day depend on materials we almost never see. Lithium and cobalt are used in the batteries of phones and electric cars. Neodymium helps motors turn in electric vehicles and wind turbines. Copper carries power through homes, factories, and data centres.
This group of metals are called critical minerals because modern economies cannot function without them. Yet, their supply is difficult to secure. Critical minerals are mined in only a few parts of the world and move through supply chains that are shaped by politics and trade policies. The real challenge, however, lies in the journey that takes place between the material leaving the mine and becoming a finished product.
That journey is where we lose our oversight. Material is visible when it leaves a mine. The finished product is visible to the customer upon arrival. But between those two points, the material passes through refineries, processing plants, component makers, and assemblers. It changes form. It crosses borders and becomes more invisible as it gets embedded inside components and across assembly units. This creates a highly complex and interconnected network that is hard to see, and this is where materials become difficult to trace.
The missing layer behind sustainability ambitions
This matters for several reasons. Firstly, a circular economy relies on materials being recovered and reused, which can only happen in a system where they are traceable and visible. Secondly, minerals are not the only thing travelling through these networks. Risk travels with them. A disruption anywhere along that path can ripple through suppliers and countries long before anyone realises they are exposed. If the path in that network is unseen, so is the risk moving through it.
The importance of this network visibility became evident in 2021, when a global semiconductor shortage forced car factories to slow down and stop production. Reports estimated that the automotive industry lost 7.7 million vehicles and around $210 billion in revenue that year.[1] The chip was the visible shortage, but it exposed a much bigger problem. An entire industry can fail because of a small component hidden in a finished product several tiers deep in a supply chain that cannot be traced.
In October 2025, after President Trump introduced wide-ranging global trade tariffs, China announced an export control on rare-earth metals that required exporters to obtain licenses before shipping seven out of 17 medium and heavy rare-earth elements. The controls were later suspended until November 2026[2] but an important point had already been made. Manufacturers around the world were forced to face a question they should have already known the answer to: which of our products depend on rare-earth metals? Which of our direct suppliers depend on that material? And where does this material come from? Many could not answer.
Policymakers have started taking this challenge more seriously. The European Union's Critical Raw Materials Act, in force since 2024, sets a target of 25% of strategic raw materials to be sourced from recycling by 2030.[3] The G7's 2025 critical minerals plan called traceability a necessary measure and urged the development of shared tracing systems.[4] These initiatives indicate that the world is heading in the right direction, but they all have the same assumption: that the path of how a material travels through the network can be seen, when in many supply chains it still cannot.
This missing layer connects circular economy, ethical sourcing, and supply chain risk management. For the circular economy, the question is whether materials can be traced inside products. In responsible sourcing, it is about verifying claims about the origin of certain materials. For supply resilience, the question is whether companies and countries can see a dependency before a disruption arrives.
We can’t recycle what we cannot see
Recycling is often seen as a promising solution to the critical mineral shortage, but it cannot work alone. Facilities can only recover the materials they receive. Governments may set recycling targets, but they cannot mandate the recycling of a magnet that cannot be located, separated, or identified at a product’s end of life. More importantly, they cannot determine which materials are present in which products, or in what composition and quantity.
Rare earth metals are a good example. Around 1% of rare earth elements are recovered from end-of-life products today.[5] The rest is lost inside motors, hard drives, speakers, screens, and industrial equipment that were never designed to be traced as material stocks. The International Energy Agency (IEA) notes in its 2024 recycling study that end-of-life permanent magnet recycling is constrained by low collection rates of below 15%, as well as difficult economic complexities.[6]
This is not a small technical gap. The study also found that stronger recycling could reduce the need for new mine development by around 25% to 40% for some key energy-transition minerals by 2050. The market value of recycled energy-transition minerals could grow fivefold to about $200 billion by 2050. But that value will not appear because recycling plants are built, it will appear when products, components, and materials can be identified, collected, and returned to the system.
The same blind spot weakens responsible sourcing. Most ethical sourcing and net-zero claims assume that a company knows where its materials came from. Most of these claims are based on first tier suppliers only. Once we go deeper in the network, that visibility becomes less clear. A company may be able to name its direct supplier, but not the refiner behind it, nor the mine behind the refiner. In this setting, a claim about responsible and ethical sourcing can be stated more easily than verified.
Visibility comes before traceability
Visibility and traceability are often used as if they are the same thing; they are not. A company has visibility when it can see the structure of its supply network: the firms and relationships that sit beyond the first supplier. But it only has traceability when it can follow a material through that structure, from origin to transformation to end use. The second depends on the first. Without the network, the material has no route to follow.
That is where the problem begins. If a company cannot see its extended supply chain network, it cannot see where a disruption hits, which companies it affects, and what dependencies it may be exposed to. In one global survey of 408 organisations across 64 countries, about a third of disruptions were traced to second-tier suppliers or beyond, while 69% of firms said they did not have full visibility of their own supply chains.[7] In other words, many companies are being disrupted by parts of the chain they do not properly see. And what they cannot see, they cannot control. The chain must become visible before the material can be traced through it.
Building the missing layer
For a long time, building this layer was labour-intensive and manual. Questionnaires were sent by companies downstream about supplier names, product information, and countries they source from were sent to suppliers who had little to no incentive to respond. As expected, the responses came back slowly and were incomplete. Most suppliers did not have the required information, while some did but treated it as commercially sensitive.
Recent advances in AI have introduced new ways to solve this long-standing challenge. Wanting visibility was never the hard part; getting the information from suppliers and keeping it up to date was the most challenging part. A new class of generative AI, referred to as agentic AI, can now do this manual work that once defeated analysts, autonomously. Unlike a chatbot that responds to a prompt and stops, an agentic system can be given a goal and left to pursue it, reading thousands of documents and following a thread from one supplier to the next, building a rich network of knowledge of the extended supply chain.
Large language models, a subclass of AI, can now analyse unstructured public text to identify companies, their products, locations, and business relationships. These relationships are mapped in a knowledge graph, connecting suppliers as part of a broader network. This approach creates a connected view of dependencies among suppliers, materials, products, and countries. Together, these methods help build multi-tier supplier and product network maps from public sources, revealing hidden dependencies.[8], [9], [10]
Public-source maps are not perfect. Their completeness and accuracy depend on the quality of available information. Some parts of the network will remain ambiguous, and others may change before public records are updated. However, for the first time, building this missing layer is becoming achievable. Companies can now construct, test, and continuously update.
Traceability should be a shared infrastructure
No single company can build this missing layer alone. If each company creates its own private map, the result will be incomplete, fragmented, and costly. While this may help individual firms manage some risks, it will not provide the comprehensive visibility and traceability needed for effective decision-making.
Traceability should be treated like a shared infrastructure rather than a private database. It becomes more useful when it is interoperable, trusted, and built to common standards by those involved. The EU’s digital product passport is beginning to implement this concept. Starting with batteries from 2027 onwards, the logic is that products should carry a digital record of key information across their life cycle.[11] The G7 roadmap is also progressing towards a shared common framework rather than a hundred incompatible, siloed, and fragmented systems.
This is where policymakers, industry leaders, and technology experts must collaborate. Policy can define the information required for traceability. Industry can make this information accessible and share it across companies, countries, and procurement and compliance teams. AI can help address gaps left and reconstruct the visibility and traceability layer. As materials enter components, cross borders, or change ownership, they can remain both visible and traceable.
Building sight back in
For most of industrial history, traceability was not a priority. Once a material became part of a product, people stopped asking where it came from. There was always more material available to mine, so there was less urgency around knowing the origin. That is changing. Critical minerals are harder to secure, and their journey matters more than ever before. And it is not just in the critical minerals sector where this matters. Food, farming, fashion, and other industries will face the same pressure, especially where people need to trust what they use and where it comes from.
Countries and companies that build this missing layer will identify exposures sooner. For governments, this means understanding which suppliers and countries their industries depend on and anticipating how external decisions could impact critical sectors. For companies, it means identifying where alternatives are needed before disruptions occur.
Traceability is now recognised as more than a sustainability tool; it is vital for national resilience. Those who understand their dependencies will not be immune to disruptions, but they will be less affected.
The opinions expressed in this article are solely those of the author and not Frontier25.
About the author: Sara AlMahri is a PhD candidate in the Department of Engineering at the University of Cambridge, working at the intersection of agentic AI systems, supply chains, and manufacturing. She is also the founder and CEO of Mina AI, a Cambridge-based startup building an AI-native workforce for enterprise operations.
