0 cumulative citations
View corpus contextReducing compute's carbon footprint requires more than engineering tweaks: accurate measurement, efficient hardware and software, circular supply chains and demand management must work together, and incentives and system-level trade-offs will determine real-world emissions reductions.
Proceedings of the 2nd International Workshop on Low Carbon Computing (LOCO 2026)
arxiv
review_meta
n/a evidence
7/10 relevance
Summary only
summary available; pdf_status=not_found
Source
PDF
Meaningful cuts in computing-related greenhouse-gas emissions require a portfolio of coordinated actions — rigorous measurement, energy- and resource-efficient software and hardware, lifecycle and circular practices, carbon-aware operations, and behavioral/sufficiency measures — rather than a single technical fix.
Citation observations
Cumulative provider counts captured on specific dates; providers are never combined.
This volume contains the proceedings of the 2nd International Workshop on Low Carbon Computing (LOCO 2026), held at Lancaster University, United Kingdom, on 10-11 September 2026. LOCO provides an interdisciplinary forum for research, practical tools, early-stage work, radical ideas, and critical perspectives addressing the reduction of greenhouse gas emissions associated with computing. The proceedings cover topics including carbon measurement and reporting, sustainable software engineering, energy-efficient computing, carbon-aware systems, hardware longevity, embodied carbon, circular computing, resource management, frugal and sufficiency-oriented computing, sustainable artificial intelligence, scientific computing, and the wider environmental effects of digital technologies. Full workshop papers were evaluated through a non-blind peer-review process by members of the LOCO 2026 Programme Committee. Submissions were assessed for originality, relevance to the workshop, quality, and potential to stimulate constructive discussion. The volume also includes accepted lightning-talk abstracts where authors elected to include them.
Summary
Main Finding
The LOCO 2026 proceedings consolidate interdisciplinary research and practical work showing that meaningful reductions in greenhouse-gas emissions from computing require a portfolio of approaches: accurate carbon measurement and reporting, energy- and resource-efficient software and hardware design, lifecycle (embodied) thinking and circular practices, carbon-aware operational systems, and behavioral and sufficiency-oriented interventions. Progress depends on measurement, incentives, and system-level trade-offs rather than single technical fixes.
Key Points
- Scope: Topics include carbon measurement & reporting, sustainable software engineering, energy-efficient computing, carbon-aware systems, hardware longevity, embodied carbon, circular computing, resource management, frugal/sufficiency-oriented computing, sustainable AI, scientific computing, and broader environmental effects of digital technologies.
- Interdisciplinarity: The workshop emphasizes technical, lifecycle, social, and policy perspectives and aims to bridge measurement, engineering, and critical/early-stage ideas.
- Lifecycle focus: Embodied carbon and hardware longevity are highlighted alongside operational energy — interventions across manufacturing, use, and end-of-life matter.
- Systems & software: Energy-efficient algorithms, carbon-aware scheduling, and resource-management techniques are promoted as levers for emission reductions in production systems and cloud environments.
- Sufficiency and circularity: Beyond efficiency, the proceedings foreground frugality, reuse, repair, and circular-economy practices as necessary complements to technical optimization.
- Evaluation & discourse: Submissions were peer-reviewed (non-blind) for originality, relevance, quality, and potential to stimulate constructive discussion; the volume includes full papers and lightning-talk abstracts.
Data & Methods
- Review & selection: Full workshop papers were evaluated through a non-blind peer-review process by the LOCO 2026 Programme Committee using criteria of originality, workshop relevance, quality, and capacity to stimulate discussion.
- Typical empirical approaches represented (inferred from topics): carbon accounting and measurement studies using energy and grid-carbon-intensity data; lifecycle and embodied carbon analyses; experimental and systems-evaluation work (e.g., benchmarking energy or latency trade-offs); design and implementation of carbon-aware schedulers or software tools; case studies of hardware longevity and circular practices.
- Formats included: full workshop papers (peer-reviewed) and accepted lightning-talk abstracts (optional inclusion).
Implications for AI Economics
- Cost structure and externalities
- Incorporate operational and embodied carbon into unit-cost models for compute (e.g., $/CPU-hour + embodied CO2 per chip). This affects true social cost and firms' investment decisions.
- Model trade-offs between performance, energy use, and emissions explicitly; optimal allocation of compute may shift when carbon is priced or regulated.
- Measurement & pricing incentives
- Accurate measurement (energy metering, grid carbon intensity, lifecycle emissions) is foundational; economists should use or help design standardized metrics for per-task or per-model emissions.
- Carbon-aware pricing (time-varying compute prices tied to grid carbon intensity) can internalize environmental costs and shift demand — evaluate welfare and distributional effects.
- Investment and innovation
- Expect demand for energy-efficient hardware, algorithmic efficiency, and circular business models; consider effects on capital allocation, technology adoption speeds, and industry structure.
- Public R&D and procurement policies aimed at embodied-carbon reduction and hardware longevity can alter long-run supply-side costs of low-carbon compute.
- Rebound and sufficiency
- Efficiency improvements can cause rebound (increased usage); sufficiency-oriented interventions and usage caps or pricing are necessary complements. Empirical work should measure rebound elasticities specific to digital services and AI workloads.
- Policy & market design
- Design of regulation (e.g., disclosure requirements, carbon accounting standards for digital services, extended producer responsibility for hardware) will shape firm behavior; analyze compliance costs and market responses.
- Consider interactions with carbon markets and whether compute-sector offsets should be allowed, and how to verify embodied-carbon claims.
- Research directions & methods for economists
- Data collection priorities: high-resolution energy and usage logs, grid carbon-intensity traces, procurement and manufacturing lifecycle data, device longevity and secondary-market prices.
- Suggested empirical designs: natural experiments exploiting time-varying grid carbon intensity or pricing changes; differences-in-differences around policy interventions or procurement rules; randomized trials of carbon-aware scheduling or price signals; lifecycle cost-benefit analyses.
- Modeling: extend production-function models of digital services to include energy and embodied-carbon inputs; incorporate dynamic adoption and vintage capital (hardware longevity, secondary markets); include behavioral demand responses to price or carbon labels.
- Practical takeaways for stakeholders
- Firms: adopt standardized measurement, prioritize software and model efficiency alongside hardware choices, consider circular procurement and longer asset life to lower total emissions and risks.
- Policymakers: require transparent carbon reporting for data centers and cloud services, incentivize embodied-carbon reductions and refurbishment markets, and evaluate carbon-aware demand management tools.
- Researchers: focus on quantifying full lifecycle emissions per unit of compute and empirically estimating demand elasticities and rebound effects to inform policy and firm strategy.
If you want, I can: (a) outline a data collection plan for estimating lifecycle carbon per AI training run, (b) draft empirical strategies to measure rebound effects for AI compute, or (c) extract potential paper topics from the LOCO 2026 themes relevant to economic research. Which would you prefer?
Assessment
Paper Typereview_meta
Evidence Strengthn/a — This is a workshop proceedings summary and research agenda synthesizing heterogeneous contributions (measurement studies, lifecycle analyses, system benchmarks, case studies) rather than reporting a unified empirical causal finding; no single identification strategy or causal estimate is presented.
Methods Rigorn/a — The text summarizes many different papers and approaches rather than applying a single, rigorous empirical method; individual contributions may vary in rigor, but the proceedings overview itself does not present a standardized research design or identification.
SampleLOCO 2026 workshop submissions: a mix of peer-reviewed (non-blind) full papers and lightning-talk abstracts covering carbon measurement and accounting (energy metering, grid-carbon intensity), lifecycle/embodied carbon analyses, systems experiments and benchmarks (energy/latency trade-offs), designs for carbon-aware schedulers/tools, and case studies on hardware longevity and circular practices; no single dataset or unified empirical sample reported in the summary.
Themesgovernance productivity
GeneralizabilityHeterogeneous, discipline-specific contributions limit ability to generalize across contexts or derive a single causal effect., Geographic and grid-carbon contexts likely vary across studies (results may not transfer between regions with different electricity mixes)., Industry data availability (e.g., cloud provider telemetry, manufacturing lifecycle data) is uneven; findings may overrepresent settings with better measurement access., Many contributions are conceptual, early-stage, or case studies rather than broadly representative empirical analyses.
Claims (8)
| Claim | Direction | Outcome | Confidence & Evidence | Details |
|---|---|---|---|---|
| Meaningful reductions in greenhouse-gas emissions from computing require a portfolio of approaches rather than a single technical fix. Other | positive | Greenhouse-gas emissions from computing |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Accurate carbon measurement and reporting are foundational for reducing the environmental impact of computing. Governance And Regulation | positive | Accuracy and availability of computing carbon measurements |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Embodied carbon and hardware longevity should be considered alongside operational energy when evaluating computing emissions. Other | positive | Total lifecycle emissions of computing hardware and systems |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Energy-efficient algorithms, carbon-aware scheduling, and resource-management techniques are potential levers for reducing emissions in production systems and cloud environments. Organizational Efficiency | positive | Operational emissions from production and cloud computing systems |
Reading fidelity
high
Study strength
low
|
not reported
|
| Efficiency improvements alone may be insufficient because rebound effects can increase usage; sufficiency-oriented interventions, usage caps, or pricing may be needed as complements. Other | mixed | Computing usage and emissions following efficiency improvements |
Reading fidelity
high
Study strength
low
|
not reported
|
| Reuse, repair, circular procurement, and longer asset lifetimes are presented as necessary complements to technical efficiency improvements. Other | positive | Lifecycle resource use and emissions from computing equipment |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Carbon-aware pricing tied to grid carbon intensity could internalize environmental costs and shift demand for computing, but its welfare and distributional effects require evaluation. Consumer Welfare | positive | Compute demand and associated environmental costs |
Reading fidelity
high
Study strength
speculative
|
not reported
|
| The workshop uses a non-blind peer-review process to evaluate full papers according to originality, relevance, quality, and capacity to stimulate discussion. Governance And Regulation | positive | Quality and selection of workshop submissions |
Reading fidelity
high
Study strength
high
|
not reported
|