Project Overview: This PhD opportunity, hosted at University College London and administered by University of Reading's AI-INTERVENE department, tackles the urgent global challenge of biodiversity decline. The project leverages advanced computational approaches, including large language models (LLMs) and machine learning, to address critical data limitations in biodiversity assessment. By developing sophisticated extraction algorithms, the research aims to mobilize hidden biodiversity records from diverse sources such as non-English scientific literature, grey literature, and corporate documents. Research Focus: The project will create comprehensive temporal biodiversity baselines spanning 1800 to 2100, using multilingual LLM extraction algorithms to identify, validate, and contextualize biodiversity records across linguistic and geographical contexts. These data will support predictive models of wildlife abundance in response to land-use, climate, and management changes, utilizing gradient boosting techniques (e.g., LightGBM, XGBoost) to enhance model accuracy and identify species and regions in decline. Innovation & Impact: By unlocking previously inaccessible scientific information with AI tools and integrating these into predictive models, the research will generate unprecedented insights into long-term biodiversity trends. The project aims to reduce linguistic and geographical biases, contributing to a more equitable understanding of global biodiversity change. Outcomes will directly improve the Living Planet Index, a key global indicator for tracking wildlife abundance. Training & Development: The successful candidate will benefit from a comprehensive training programme in applied AI, biodiversity, and transferable research skills. A placement with an AI-INTERVENE project partner (3-18 months) is included, along with opportunities to present at national and international conferences, positioning the student at the forefront of the discipline and enhancing future career prospects. Funding: The studentship is fully funded by the AI-INTERVENE NERC Doctoral Focal Award, subject to a competitive selection process for the strongest applicants. Eligibility: Applicants should have strong computational skills, experience with machine learning, natural language processing, and programming in R or Python. Experience with large datasets and predictive statistical modelling is highly desirable. The project welcomes candidates from diverse academic backgrounds. Application Deadline: January 19, 2026. How to Apply: Apply via the FindAPhD project link. Prepare your application materials emphasizing relevant computational and data science experience. Selection is competitive; contact the department for further details if required. References: Key literature includes works by Cornford et al. (2021, 2023), Ledger et al. (2023), and the WWF Living Planet Report (2024).