In South Asia, a region facing rapid economic growth, immense population pressure, and high climate vulnerability, the circular economy (CE) has become a critical imperative for sustainable development. This study provides a comparative overview of the CE landscape across eight South Asian countries: Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka. The analysis reveals the CE transition is nascent region-wide, though India has advanced its policy landscape through a comprehensive suite of rules and missions and Pakistan is developing a national policy. The primary focus remains on waste management, evidenced by programs like Bhutan’s ‘Zero Waste by 2030’ vision, the Maldives’ Single-Use Plastic Phase-Out Plan, and Sri Lanka’s Clean Sri Lanka Programme. While Extended Producer Responsibility (EPR) is emerging for plastics and e-waste in India, Bangladesh, Sri Lanka, and Pakistan, a significant “policy-practice gap” persists, undermined by weak enforcement and governance fragmented across priority sectors like plastics, food systems, and textiles. Most major CE initiatives are catalyzed by international development partners, with regional programs playing a key role in funding innovation. Finally, while the informal sector is the backbone of material recovery, ensuring a just transition that improves working conditions and secures livelihoods remains a critical challenge. The absence of a cohesive regional framework limits collaboration. Scaling the circular economy in South Asia requires integrated national strategies, prioritizing a just transition for the informal sector, and establishing a regional platform for policy harmonization to create self-sustaining system through multi-sectoral involvement, including the business sector.

Efforts to scale renewable electricity in semiconductor manufacturing are not keeping pace with AI-driven growth in electricity demand. This report examines renewable electricity strategies in the tech and semiconductor sectors and sets out recommendations for closing the gap. Tech companies' electricity demand is growing at unprecedented rates, as the rapid expansion of artificial intelligence (AI) and associated data centre build-out drives a stark rise in electricity consumption and heightens pressure on energy systems worldwide. While public debate tends to focus on the impact of data centres and their electricity consumption, upstream supply chains account for at least half of major tech companies' total greenhouse gas (GHG) footprint. Semiconductor manufacturing is a particularly important emissions hotspot within these supply chains: it is highly electricity-intensive, and chips are essential components in smartphones, laptops, data centre hardware and AI accelerators. For fabless companies such as NVIDIA, which design but do not manufacture chips, and for companies that outsource production or purchase large volumes of servers and devices, such as Apple, Google, Amazon, Meta and Microsoft, these emissions sit upstream in the value chain, embedded in purchased goods and capital equipment. Transitioning semiconductor production to renewable electricity is therefore essential for reducing both semiconductor manufacturers' own emissions and the supply chain emissions of major tech companies. Drawing on company assessments, stakeholder interviews and desk research, this report examines the barriers to this transition and the actions needed from semiconductor manufacturers, major tech companies, standard-setters and policymakers. Why does this matter now? South Korea and Taiwan are central to this transition, but renewable electricity supply in both markets is not keeping pace with growing semiconductor demand. Both countries are positioning themselves as key drivers of the global AI boom, exemplified by South Korea's launch in June 2026 of the three new chip and AI 'mega projects' covering semiconductors, physical AI and AI data centres. South Korean firms account for over 60% of global memory chip production, while Taiwanese firms account for more than 90% of advanced chip manufacturing, and demand from both industries is projected to grow further. Renewables supplied only around 10% of electricity in South Korea and 12% in Taiwan in 2025, and national renewable electricity targets remain relatively unambitious given expected demand growth. Both governments are also planning additional fossil gas capacity to serve new industrial demand, deepening dependence on imported fossil fuels and heightening vulnerability to price volatility, while making corporate and national climate targets harder to achieve. Long lead times for new nuclear capacity mean it cannot substitute for the rapid renewable electricity expansion needed in the near term. What are our main insights? Corporate renewable electricity procurement remains difficult in both countries, although the main barriers differ. Taiwan's challenge is primarily the limited volume of renewable electricity available for companies to contract and claim, compounded by grid constraints, permitting delays and matching and settlement rules. South Korea's challenge is primarily high project costs, grid congestion and curtailment risk, with its electricity market directing most corporate demand towards the Green Premium, an accessible but low-additionality procurement option. In both markets, long-term power purchase agreements (PPAs) linked to new projects can provide stronger support for additional renewable capacity but remain more costly and difficult to access. All semiconductor manufacturers assessed except Micron have committed to reach 100% renewable electricity by 2040 or 2050, but these timelines lag behind power-sector decarbonisation benchmarks, and several manufacturers lack ambitious interim milestones. Renewable electricity uptake remains low in South Korea and Taiwan, where a significant share of production takes place. South Korean manufacturers continue to rely heavily on the Green Premium. SK Hynix has already built gas-fired power plants near its fabrication facilities, and both SK Hynix and Samsung are planning additional on-site gas-fired capacity, risking long-term dependence on fossil gas. Major tech companies could accelerate renewable electricity uptake in semiconductor supply chains by setting clear targets and helping manufacturers overcome market barriers, but most are not yet doing so. Customer demand has been an important driver of procurement, though buyer pressure may carry less weight amid the current chip shortage. Apple is the only major tech company assessed with a formal supply chain renewable electricity target; Google and Microsoft send relatively strong demand signals through requirements for suppliers to source renewable or carbon-free electricity by 2030, while Amazon, AMD, Meta, NVIDIA provide weaker signals or no specific requirements at all. Only a few companies go beyond setting targets and supplier requirements to actively help suppliers overcome procurement barriers. Capacity-building activities are relatively widespread, but sustained technical and direct financial support remain far less common. Apple's Supplier Clean Energy Programme and China Clean Energy Fund, and TSMC's collaborative PPA with its suppliers, stand out as examples of how large buyers can help suppliers overcome market barriers and could be replicated more widely across the sector. What are our key recommendations? What should semiconductor manufacturers do? Adopt ambitious emissions reduction and renewable electricity targets Shift to renewable electricity procurement that supports new capacity Use purchasing power to support suppliers in accessing renewable electricity Advocate publicly for renewable electricity market reform What should major tech companies do? Adopt ambitious supply chain renewable electricity targets Actively support suppliers in sourcing renewable electricity Publicly advocate for renewable electricity market reform in manufacturing regions Strengthen transparency on supply chain renewable electricity procurement What should policymakers prioritise? Both the South Korean and Taiwanese government should accelerate renewable electricity build-out and grid infrastructure upgrades For a detailed, side-by-side comparison of the renewable electricity procurement options available to companies in Taiwan and South Korea, download the full overview (here).

Future AI growth will increase demand for Australian minerals and energy Australian mining a major user of AI to boost productivity and reduce costs AI’s success in Australia should be based on technology-neutral and evidence-based regulation. Australian mining is both an advanced user of artificial intelligence and crucial to its growth and ability to reshape our economy by providing the energy and critical minerals required for data centres and other AI infrastructure. The potential benefits of efficient AI adoption in Australia are large; the Minerals Council of Australia estimates in its submission to the Joint Select Committee on Artificial Intelligence that AI’s impact on total factor productivity could increase Australia’s GDP by at least $45 billion by 2030. As demand for AI grows, so too will demand for the minerals and energy required to build and power it. AI systems depend on data centres, advanced electronics, communications networks, electricity generation and transmission infrastructure. Expanding this infrastructure will drive significant additional demand for a range of critical minerals of which Australia is already a reliable producer, including: Iron ore and metallurgical coal for construction and energy Copper for data centres, grids, generation, storage and communications Uranium as a potential input to the reliable electricity systems supporting AI-related loads. The opportunity is substantial: the International Energy Agency projects that data centre electricity consumption will more than double from 2024 to about 945 terawatt hours by 2030, exceeding Japan’s total electricity consumption. Australia’s major miners and smaller producers are well placed to benefit. For example, Rio Tinto estimates that it produces up to 70 per cent of the raw materials required for each gigawatt of data centre capacity. Australian mining is doing more than supplying the foundations of AI: it’s also helping to lead its adoption. The submission outlines current uses of AI in the Australian minerals industry including: BHP’s Olympic Dam operation uses AI and machine learning models to monitor the condition of haul trucks, crushers, pumps and other critical plant equipment. Sensors collect real-time data on vibration, temperature, and operational parameters, which AI algorithms analyse to predict potential equipment failures before they happen CSIRO, in partnership with Rio Tinto, has developed advanced geospatial AI techniques to analyse satellite imagery and geological datasets to identify promising mineral deposits across Western Australia. By integrating AI with remote sensing data, the project has enhanced the accuracy and speed of orebody mapping, reducing exploration costs and environmental impacts. These practical examples show how Australian mining companies are already using AI to solve real-world challenges in complex, safety-critical environments. The submission includes recommendations to encourage the use of AI while providing appropriate safeguards such as: A technology-neutral and evidence-based approach to AI regulation Any additional regulation to be targeted at new identifiable harms while avoiding unnecessary barriers to innovation, investment and productivity growth Accelerating opportunities to use AI to improve regulatory performance including AI-enabled approvals systems which have the potential to improve the timeliness and quality of regulatory decision-making while maintaining appropriate human oversight and accountability Developing a larger pool of technical and trade skills to support the continued development and adoption of AI across the economy.

Restoring confidence in Australia’s minerals sector regulation Cultural heritage protections can’t be exploited by taxpayer-funded anti-mining activists Responsible mining and heritage protection not mutually exclusive. Today’s Federal Court decision setting aside the Commonwealth’s 2024 cultural heritage declaration affecting Regis Resources’ McPhillamys Gold Project in New South Wales helps to restore confidence in Australia’s regulatory processes and the integrity of decisions supporting Australia’s ability to be a magnet for major mining investment. Regis Resources secured NSW and Commonwealth environmental approvals following extensive assessment and engagement with the Orange Local Aboriginal Land Council, which is recognised as the local Aboriginal representative body. The subsequent late-stage intervention by the Commonwealth put more than $1 billion in proposed investment at risk and raised serious questions about the reliability of Australia’s approvals framework. Australian mining investment relies on certainty of process, even if the outcomes of that process are not certain. Support for responsible mining and protecting Aboriginal cultural heritage are not competing or mutually exclusive objectives. Both depend on transparent decision-making, credible evidence and meaningful recognition of the appropriate Traditional Owner representatives. The McPhillamys experience has exposed significant weaknesses in how competing claims to cultural authority are considered, particularly where the views of recognised local Aboriginal leaders and Elders differ from those without recognised spiritual connection to Country. Cultural heritage protections must not be exploited by taxpayer-funded anti-mining activist organisations, including the Environmental Defenders Office, to undermine legitimate Aboriginal representation and obstruct responsible development. The livelihoods of regional communities, Aboriginal businesses and workers depend on a balanced framework that genuinely protects cultural heritage, respects the voices of Traditional Owners and gives investors confidence that projects will be considered under established regulatory processes. The MCA will continue to work with the Federal Government, the First Nations Heritage Protection Alliance and others to address the weaknesses exposed by the Federal Court today.

Supporting early-stage industry development of downstream antimony processing in Victoria Victoria in top ten of global antimony producers Opportunity to lead in minerals processing technology The announcement of grants to Gekko Systems and Southern Cross Gold Consolidated Ltd today by the Victorian Government is a crucial first step in securing an antimony processing facility in Victoria. Developing homegrown capacity to process our critical minerals adds value, creates jobs and secures our supply chain. Antimony is one of the Australian Government’s top three priority critical minerals. Its strategic role in defence and energy technologies including photovoltaic glass, batteries, armoured vehicles and as a flame retardant, coupled with limited global supply and few substitutes, make it a critical mineral. Australia has the world’s third-largest reserves of antimony and Victoria is in the top ten producers of antimony globally, making the state a logical choice for antimony processing close to existing and future mines. The grants help Victoria build on its competitive advantages in education, research and manufacturing strengths to be a leader in minerals processing technology. Backing research and pilot plants combined with reducing environmental impacts supports the state’s circular economy goals and Victoria’s expertise in education, METS and manufacturing. The next step is for whoever forms government after next month’s Victorian election to ensure competitive tax, energy costs and permitting timeframes. Along with expanded investment in mining and processing, the right policies will support Victoria’s development into a global hub for antimony and the production of mineral sands and rare earths.

A team of scientists in Norway has found evidence for a new, previously unknown mechanism for the development of cancer in children: they seem to produce an unusual amount of aberrant antibodies against the building blocks that make up the DNA repair machinery, a diverse set of hundreds of proteins which identify and correct damage or errors in the genome. This important discovery could open the door to the early screening of small children, in the hope of ultimately preventing them from developing such pediatric cancers years in their teenage years. “We showed that many children with various types of cancer have antibodies against some of their own DNA repair proteins. We also demonstrated that such ‘auto-antibodies’ can enter live cells and bind to – and probably inhibit – the activity of these essential proteins,” said corresponding author Dr Ola Myklebost, an emeritus professor of molecular oncology at the University of Bergen. The authors published their findings in Frontiers in Immunology. Why pediatric cancers are different Pediatric cancers, such as leukemia, myeloma, neuroblastoma, and osteosarcoma, differ from adult cancers in their causes, location and type, and response to therapy. Compared to adults, children have had less time to accumulate cancerous DNA damage through exposure to environmental stressors, lifestyle factors, and aging. Pediatric cancers thus tend to carry fewer mutations than adult ones and arise from less mature cells, for example stem or progenitor cells. Serguienko, Myklebost, and colleagues characterized the ‘proteome’ – the ensemble of expressed proteins – in tumor and blood samples from 175 adults and children with sarcoma or acute lymphoblastic leukemia, collected between 2014 and 2017 through the NoSarC Norwegian Sarcoma Study, the Norwegian Childhood Cancer Biobank, and the Oslo University Hospital Cancer Research Biobank. They compared these with samples from healthy anonymized people such as blood donors and children participating in the Bergen Growth Study or visiting a clinic in Bergen. The results showed that compared to young adults and older people, children between two and 19 years of age with pediatric cancer had much greater numbers antibodies of the IgG class against DNA repair proteins circulating in their blood. These auto-antibodies mainly targeted RAD52, NUMA1, ALKBH7, APBB1, UNG, and MGMT, all key genes closely involved in DNA repair, but not significantly mutated in these cancers. Could these auto-antibodies explain why these children had developed cancer? To find out, the scientists looked at selected samples from 59 school age children who had taken part in the British ‘Avon Longitudinal Study of Parents and Children’ and ‘Born in Bradford’ long-term cohort studies. Half of these children had developed cancer years later (but before the age of 30 years), while the other half had not. “Although this cohort is small, it was the best we could find internationally,” said Myklebost. “Even so, we were able to show that indeed, children who later got cancer (on average six years after blood sampling) were about eight times more likely than their healthy peers to have antibodies to at least one core antigen involved in DNA repair.” Likely role in development of cancer The authors hypothesized that the targeting of DNA repair proteins by these auto-antibodies could provide a phenotype resembling what we see for related proteins with defective hereditary variants that cause a predisposition to cancer. When they enter a cell or its nucleus, they could bind to and immobilize or inhibit directly the targeted proteins and thus hinder the normal DNA repair. Such weakened repair might in turn cause pediatric cancer. Crucially, in follow-up experiments the scientists were able to show that these auto-antibodies can enter cells, especially during the phase when cells end their growth and prepare to divide. Inside the cell, they seem to zone in on the TIGER organelle, a site where transcription factors and other short-lived but essential proteins are produced and modified. “[These results] together provide a plausible mechanism for increased cellular damage that could lead to cancer development,” concluded the authors. But what might trigger the production of antibodies against normal DNA repair proteins? “They could be triggered by infections, by crossreacting to some microbial proteins or overstimulation of the immune response, or by some imbalance in immune tolerance,” hypothesized Myklebost. The results immediately point to the possibility of screening children to detect any autoimmune predisposition to cancer early, with the hope of preventing their later progression to the disease or its early detection. “At present we are investigating the effect of patient antibodies on the repair of induced mutations in cultured cells. Ultimately, we hope to clone white blood cells that secrete such auto-antibodies, and test for any cancer-causing effects in mice to definitively prove our hypothesis,” said Myklebost. Journal Frontiers in Immunology DOI 10.3389/fimmu.2026.1911932 Method of Research Experimental study Subject of Research People Article Title Pathogenic autoantibodies targeting DNA repair proteins are shared by cancer patients and healthy children later diagnosed with cancer Article Publication Date 7-Oct-2026 COI Statement The authors declare no competing interests, except that AS and OM have submitted a patent application for the use of core autoantibodies as biomarkers