# Ezequiel Lopez-Lopez

> Computational social scientist at TU Dresden (SynoSys) and Guest Researcher at the Max Planck Institute for Human Development, Berlin. Research focus: AI-assisted qualitative research, societal impacts of generative AI, and policy implications.

## Affiliations

- **TU Dresden — SynoSys (Center Synergy of Systems)** — Research Associate
- **Max Planck Institute for Human Development, Berlin** — Guest Researcher
- Freelance AI consultant

## Background

- PhD in Computer Science, TU Berlin (expected 2026)
- MSc in Language Technologies, UNED Madrid (2020) — specialisation in NLP & IR
- BSc in Physics, University of Granada (2016) — computational + mathematical-physics focus
- Prior industry: AI & Data Consultant at IBM Spain (2016–2018) and IBM Germany (2018–2020); senior AI/NLP engineer at DB Systel (Deutsche Bahn, 2020–2022); engineering trainee at the European Space Agency on the Cluster Mission (Madrid, 2015).

## Research themes

- **AI-assisted qualitative research** — LLM-based citizen-perspective elicitation at scale; network analysis of open-ended responses; argument mining; transformer + classical-ML hybrids.
- **Societal impacts of generative AI** — observational studies of large media corpora (YouTube, podcasts, social media) for societal drifts and misinformation events; controlled behavioural experiments on confirmation bias, hypercustomization, and metacognition; LLMs used as instruments to measure LLM effects.
- **Policy implications** — Rapid Think Tanks (atomic deliberation protocols); citizen-perspective elicitation applied to the 2023 Australian Voice referendum, the Zeitenwende foreign-policy debate, the 2024 Valencia flash-floods, and cross-country opinion in six European countries; conceptual mapping of GenAI risks.

## Skills

- **Methods:** NLP & argument mining; knowledge graphs; network science; LLM-based analysis; computational social science; behavioural experiments; observational causal inference on large media corpora.
- **Programming:** Python (strong); R, Node.js, bash, C++ (working).
- **Libraries / frameworks:** spaCy, Hugging Face Transformers, scikit-learn, TensorFlow, allennlp, NetworkX, pandas, NumPy.
- **Databases & infrastructure:** Neo4j (knowledge graphs), Elasticsearch, MongoDB, MySQL, Docker, git, CI/CD; cloud: AWS, Google Cloud, Azure, IBM Cloud.
- **Spoken languages:** Spanish (native), English (C2), German (B2–C1), French (B1).
- **Domain experience:** Health, climate, electoral / referendum, cross-country policy, banking, insurance, retail, automotive, media, energy, infrastructure, public transport.

## Currently open to

Research scientist, postdoctoral, and consulting opportunities at the intersection of computational social science, AI policy, and responsible / safety-oriented AI research. Based in Berlin; open to relocation within the EU.

## Selected publications

- Burton, J. W., Lopez-Lopez, E., et al. (2024). How large language models can reshape collective intelligence. *Nature Human Behaviour*. https://doi.org/10.1038/s41562-024-01959-9
- Yakura, H.\*, Lopez-Lopez, E.\*, Brinkmann, L.\*, et al. (2024). Empirical evidence of LLMs' influence on human spoken communication. arXiv:2409.01754. https://arxiv.org/abs/2409.01754
- Lopez-Lopez, E.\*, Abels, C. M.\*, Holford, D.\*, Herzog, S. M., & Lewandowsky, S. (2025). Generative AI-mediated confirmation bias in health information seeking. *Annals of the New York Academy of Sciences*. https://doi.org/10.1111/nyas.15413
- Abels, C. M.\*, Lopez-Lopez, E.\*, et al. (2025). The governance & behavioral challenges of generative AI's hypercustomization capabilities. *Behavioral Science & Policy*. https://doi.org/10.1177/23794607251347020

## Profiles

- Google Scholar: https://scholar.google.com/citations?user=hUB_UyEAAAAJ
- ORCID: https://orcid.org/0000-0001-8709-8044
- LinkedIn: https://www.linkedin.com/in/elopezlopez

## Structured data

The full bibliography, talks, workshops, funded projects, and venn-diagram region assignments are in machine-readable JSON at https://www.ezequiellopezlopez.org/content.json