Net zero, AI and the changing landscape of copper

Satellite view of Chile’s Escondida mine. Credit: Gallo Images/Orbital Horizon/Copernicus Sentinel Data 2021 via Getty Images

In today's carbon-conscious mining world, AI and precious metals have a vitally important role to play. Instability in the energy market and a global push toward net zero by 2050 both support the same message – a need to decarbonise, lower costs and boost resilience. 

“That means a shift in the demand dynamics for a host of metals to expand renewable energy sources, evolve battery technology, and build green infrastructure,” writes Matt Britzman, equity analyst at British bank Hargreaves Lansdown.

Copper is one of these metals. Copper boasts the best thermal and electrical conductivity of any non-precious metal, making it valuable for creating efficient and reliable energy solutions. Mining giant Glencore contributes significantly to the global supply of copper, and it expects demand to outstrip supply by around 50Mt by 2030 based on the current trajectory.

To put that in context, Glencore expects total copper supply to reach around 300Mt by 2030, meaning a significant shortfall based on these projections. Copper, historically, has been deemed too common to classify as precious, but its value continues to grow, and its worth in decarbonisation is plain. With companies investing to keep supplies flowing, many are treating it as the most precious of metals.

Since the immediate threat of Covid receded, the mining industry and many others have been wary of making large multi-billion-dollar investments to bring new sites online. As costs continue to rise, large scale greenfield projects could be even more unlikely. In fact, Glencore estimates industry investment in expanding copper supply in 2025 will be some 61% below the 2012 peak.

Hence, for the bigger players, Hargreaves Lansdown expects to see more mergers and acquisitions. Rio Tinto’s work to acquire the remaining stake in Canadian mining company Turquoise Hill for $3.3bn is a prime example.

But beyond the bank’s analysis, there seems wider evidence for copper's crucial role and new ways to make the best of existing assets. A collaboration between BHP and Microsoft is using artificial intelligence (AI) and machine learning, with the aim of improving copper recovery at the world’s largest copper mine, Escondida.

Australian giant BHP operates and owns 57.5% of Chile’s Escondida mine, a joint venture with Rio Tinto (30%) and Japan-based JECO Corp (12.5%).

The company’s chief technical officer Laura Tyler has said that by augmenting new digital technology capabilities with new ways of working, the team at Escondida will be well-positioned to generate more value from an existing resource.

“We expect the next big wave in mining to come from the advanced use of digital technologies. As grades decline at existing copper mines and fewer new economic discoveries are made, next-generation technologies like AI, machine learning and data analytics will be needed to unlock more production and value from our existing mines,” she said.

In its estimates, BHP projects that the world would need to double the amount of copper produced over the next 30 years, relative to the past 30, to keep pace with the development of decarbonisation technology such as electric vehicles, offshore wind and solar farms assumed under its 1.5°C scenario.

Microsoft says Escondida produces over one million metric tonnes of copper per annum. To improve this, the company has built an AI involving BHP’s Azure Machine Learning operations platform and other Azure services such as Azure Synapse Analytics and Azure Data Lake Storage.

The hope is that these technologies will enable BHP to optimise the mine’s concentrator circuit, which is responsible for extracting, floating and collecting the copper mineral from crushed and milled ore deposits. The mine is now expanding the technology to a second concentrator at Escondida.

As the operator, BHP uses real-time plant data from the copper concentrators and machine learning systems to make hourly predictions. These predictions are then used to create machine learning–assisted recommendations for its Escondida operating team.

Consulting firm McKinsey recently released a new report emphasising the economic potential of AI. It says that AI could add the equivalent of $2.6 trillion to $4.4 trillion annually across the 63 use cases analysed. By comparison, the UK’s entire GDP in 2021 was $3.1 trillion.

Specific to workstreams that mining uses, the report also looks at the applications of generative AI, systems that can create text strings and “talk” as if human. McKinsey estimates that generative AI could enable labour productivity growth of 0.1% to 0.6% annually through to 2040, depending on the rate of technology adoption and redeployment of worker time into other activities.  

Combining generative AI with all other technologies, work automation could add 0.2 to 3.3 percentage points annually to productivity growth.

Following McKinsey's numbers points to why Microsoft and BHP are putting the hours into industrial AI. Estimates tell that Escondida produced some $10bn of copper, way back in 2007. On a turnover of billions, a 3.9% increase across automation and labour could win you some $390m.

This would come without breaking ground on new greenfield sites nor paying for pricy exploration, testing and construction for facilities. But McKinsey also strikes a cautious note: “These tools have the potential to create enormous value for the global economy at a time when it is pondering the huge costs of adapting and mitigating climate change.

“At the same time, they also have the potential to be more destabilizing than previous generations of AI. They are capable of that most human of abilities, language. This is a fundamental requirement of most work activities linked to expertise and knowledge, as well as a skill that can be used to hurt feelings, create misunderstandings, obscure truth, and incite violence.”

Mining is an industry where jobs count, often in underdeveloped or developing parts of the world. There are compelling arguments on how AI can help get metals we badly need into the global supply chain, for considerable profit. But further thoughts may also be necessary, to ensure value is also retained across mining workforces, and that the value to communities of employing a workforce is itself maintained.