{"id":14307,"date":"2024-01-22T15:49:21","date_gmt":"2024-01-22T15:49:21","guid":{"rendered":"http:\/\/TheNextWeb=1403229"},"modified":"2024-01-22T15:49:21","modified_gmt":"2024-01-22T15:49:21","slug":"meet-climategpt-an-open-source-ai-platform-fighting-climate-disinformation","status":"publish","type":"post","link":"https:\/\/www.londonchiropracter.com\/?p=14307","title":{"rendered":"Meet ClimateGPT \u2014 an open source AI platform fighting climate disinformation"},"content":{"rendered":"\n

A collective of startups<\/a> known as the Endowment for Climate Intelligence (ECI) has unveiled the first-ever open-source ensemble of AI models dedicated to addressing the impacts of climate change.  <\/span><\/p>\n

Large language models (LLMs) such as ChatGPT and Google\u2019s Bard have faced concerns over the accuracy of the information they provide, as they are trained on<\/span> data<\/span><\/a> scraped from the internet that is not rigorously fact-checked or referenced.  <\/span><\/p>\n

In contrast, ECI claims that ClimateGPT aims to create \u201ctrust and transparency\u201d by drawing on scientific data using a robust model that authenticates, secures, and governs the information provided. The model is integrated with Hedera, a public<\/span> blockchain<\/span><\/a>, which ensures the \u201chighest standards of data integrity,\u201d said the ECI. <\/span><\/p>\n

Toward open-source<\/h2>\n

ECI was formed by three startups looking to make reliable climate science information quickly and freely available to researchers, policymakers, and business leaders. <\/span><\/p>\n

\n
<\/p>\n

TNW Conference 2024 – Speakers announced!<\/p>\n

Meet the powerhouse experts that will take the stage on June 20 & 21 in Amsterdam and save your seat today.<\/p>\n

<\/a><\/div>\n<\/div>\n

Jonathan Dotan,<\/span> founder of EQTY Lab, and a founding member of ECI, believes that ClimateGPT is part of a broader industry shift toward open-source<\/a> AI. <\/span><\/p>\n

\u201cWe saw the release of countless AI models in 2023, but many still exist in walled gardens, which limits their potential for positive change,\u201d he told TNW. \u201cThis year I believe we will see demand for AI that is more open, secure, and accountable.\u201d <\/span> <\/span><\/p>\n

The data<\/h2>\n

Rather than one chatbot, ClimateGPT is an ensemble of three task-specific LLMs, that vary in terms of the number of parameters they were trained on \u2014 which also effects accuracy. For example, the 7B model is far less accurate than the 70B version. \u2018B\u2019 stands for \u2018billion\u2019 and indicates the number of data points the model was trained on. All three models are fine-tuned versions of Meta\u2019s Llama-2 large language model.  <\/span><\/p>\n

ClimateGPT was trained on data from the Dutch startup Erasmus.AI\u2019s planetary scale corpora. The Erasmus corpus is drawn from over 10  billion web pages and millions of open-access academic articles from across the natural, social, and economic sciences. ClimateGPT\u2019s skills were also honed on data from the Club of Rome\u2019s Earth4All climate science initiative and the UN Sustainable Development Goals (SDGs).  <\/span><\/p>\n

In addition, Switzerland-based startup EQTY Lab led the integration of <\/span>a new advanced cryptographic framework that authenticates, secures, and governs the model. While<\/span> AppTek, a startup specialising in machine learning<\/a>, fine-tuned the ClimateGPT through a series of tests and benchmarks. <\/span>  <\/span><\/p>\n

Countering misinformation<\/h2>\n

The chatbot synthesises interdisciplinary research and breaks silos to form a holistic understanding of the impacts of climate change. <\/span><\/p>\n

Thanks to software from AppTek, ClimateGPT is available in 20 languages. Its creators also say that it was trained and is hosted entirely on renewable energy. <\/span><\/p>\n

What\u2019s more, ClimateGPT has passed a number of tests and controls ensuring that it does not spread climate misinformation.<\/span><\/p>\n

A quick test<\/h2>\n

The first version of ClimateGPT, which is completely open-source, has been released on the Hugging Face community platform. You can also request access to it on the ECI <\/span>website<\/span><\/a>.<\/span><\/p>\n

In a quick test, we asked the chatbot to answer a basic question related to a<\/span> recent article we wrote on the problems of bioenergy<\/span><\/a>. Its answer was detailed, thorough, and hit on many of the points addressed in our article. Importantly, it was backed up by academic sources. We also liked that the LLM gave specific examples to go along with its explanations, something tools like ChatGPT or Google Bard rarely do.  <\/span><\/p>\n

Take a look for yourself:<\/span><\/p>\n

\"climateGPT