진학프로석·박사 채용 접수 플랫폼

The person hired for this position will be responsible for conducting research on the behavioral properties of large AI models (such as large language models (LLMs), with a particular focus on their persuasive capabilities, susceptibility to generating misinformation, and reproduction of toxic or biased content. Key responsibilities will include:

Designing and conducting experiments to assess the capacity of AI models to produce manipulative, misleading, or harmful textual outputs;

Developing metrics and evaluation protocols for detecting toxicity, bias, factual inconsistency, and cognitive distortion in model responses;

Investigating control mechanisms for mitigating undesired behaviors through prompt engineering, fine-tuning, model alignment, or data filtering techniques;

Analyzing multilingual and cross-cultural model behaviors in sensitive or high-risk communicative contexts;

Documenting research findings through scientific publications, internal benchmarks, and diagnostic reports;

Collaborating with interdisciplinary teams addressing ethical, societal, and regulatory aspects of generative AI systems;

Contributing to the preparation and execution of national and international research projects (e.g., NCN, Horizon Europe, UNESCO).

In addition, the person in this position may act as a scientific supervisor for a small team of junior researchers (1–2 people), supporting them in content analysis, experiment planning, and interpretation of results.