ChatNetZero, the world’s first climate target chatbot, has unveiled a major feature upgrade improving its speed, transparency, and computational efficiency as a climate accountability tool.
Originally launched in 2023, ChatNetZero was built to improve the transparency of net zero targets by enabling policymakers, journalists, investors, and civil society to scrutinise climate pledges. Unlike general-purpose large language models (LLMs), such as ChatGPT, ChatNetZero is purpose-built for climate accountability. Its outputs are grounded in peer-reviewed datasets and include source citations, reducing the risk of ‘hallucinated’ responses, that currently limit the credibility of general-purpose AI as a climate research tool.
At the same time, the rapid growth of AI is driving a surge in energy demand, pushing total data centre electricity use to around 945 TWh by 2030, slightly more than the entire consumption of Japan today. While AI has major potential to support climate action, transparency and regulatory oversight of its environmental impacts remains limited.
ChatNetZero (3.0) addresses this challenge by displaying the energy consumption of each query in watt-hours, alongside real-world equivalents, such as the power needed to run common household appliances. By disclosing this information, the platform helps policymakers, researchers, and climate professionals better quantify the true environmental impact of AI systems.
Because ChatNetZero is trained on targeted climate datasets and documents, the computational workload required for each query is significantly lower than general-purpose AI. For example, if the population of a country the size of Ireland asked the same question using ChatNetZero instead of a general-purpose AI, it could save around 1,410 kWh of electricity — roughly the amount a typical UK household uses over several months.
ChatNetZero co-developer, James Zhang of Arboretica, said:
“Transparency cannot be optional in the age of AI, especially as these systems increasingly shape public understanding, policy decisions, and investment flows.
“If AI is to fulfil its promise as a tool for tackling the climate crisis, it must also be accountable for its own environmental footprint. By making energy consumption visible, we’re showing that this level of transparency is possible and meaningful – and calling on other AI companies to follow suit.”
A Redesigned Architecture for AI Efficiency
The upgrade reflects the rapid evolution of AI over the past two and a half years. Earlier versions processed thousands of documents in a single batch using a static retrieval process. The new platform introduces a dynamic, agent-based retrieval architecture that selects the right tool and loads only the documents relevant to each user query.
Core ChatNetZero (CNZ) upgrades include:
- Updated Net Zero Tracker (NZT) Database Integration
- CNZ continues to draw on data from the Net Zero Tracker (NZT) – the world’s most comprehensive review of net zero commitments.
- The platform now synchronises continuously with the NZT database, ensuring users access the most up-to-date data and citation-backed analysis.
- Dynamic Document Intelligence
- CNZ loads only relevant documents per query, reducing unnecessary computational overhead and enabling seamless integration of new datasets.
- Per-query Energy Disclosure
- Estimates energy use using published benchmarks and response-time modelling.
- Displays energy consumption in watt-hours with contextual equivalents.
- Enhanced Mathematical Capability.
- General-purpose LLMs struggle with precise quantitative reasoning. CNZ incorporates a built-in calculation function purpose-built for climate-related data.
- This improves the platform’s ability to perform emissions comparisons, percentage reductions and target-alignment analysis. E.g. it can now provide mathematically-verified answers to questions such as:
- “How many companies in Germany have set net zero targets?” Or “What proportion of firms have a 2050 net zero target with interim milestones?”
Dr. Angel Hsu, co-developer of ChatNetZero and Director of Data-Driven EnviroLab, said:
“As large language models become more powerful, they also become more resource-intensive. We redesigned ChatNetZero to move away from monolithic document processing toward a targeted retrieval system that improves efficiency without compromising analytical integrity.
“By dynamically retrieving only relevant materials and disclosing estimated energy use per query, we are strengthening both climate accountability and AI accountability.”
Aligning with the COP30 Information Integrity Declaration
The launch comes amid growing international attention on information integrity in climate discourse. In November 2025, UNESCO and partners advanced the COP30 Information Integrity Declaration on Climate Change, calling for stronger safeguards against misinformation, enhanced transparency in digital systems, and greater accountability in AI.
Dr. Angel Hsu, Director of Data-Driven EnviroLab, added:
“ChatNetZero 3.0 shows what is possible when AI is designed with accountability at its core. ChatNetZero 3.5 will go further, giving decision-makers an AI partner that doesn’t just generate answers but investigates. We believe this is the direction the entire field of climate AI should be heading.”