New MJRC Study Evaluates Free AI Chatbots on Civic and Human Rights Across 15 Countries

The Media and Journalism Research Center (MJRC) has released a new cross-national study examining how leading free AI chatbots respond to questions about civic values and human rights. Titled Evaluating Free AI Chatbots on Civic and Human Rights Topics: A 15-Country Expert Study, the report offers an independent assessment of how widely used generative AI systems handle politically and socially sensitive issues.

As generative AI tools become embedded in everyday information-seeking, from students and journalists to policymakers and civil society actors, questions about their reliability, neutrality, and respect for fundamental rights have grown more urgent. Free chatbots in particular are often the most accessible tools globally, making their performance in different political and linguistic contexts especially consequential.

Against this backdrop, MJRC’s new report adopts a structured, expert-led methodology. Researchers in countries submitted a common set of prompts to leading free AI chatbots covering civic participation, human rights standards, media freedom, and politically sensitive themes. The responses generated by the chatbots were then systematically assessed by country experts and subject-matter specialists, who evaluated them for factual accuracy, balance, contextual understanding, and respect for fundamental rights. This expert review process allowed the study to move beyond automated scoring and provide grounded, country-specific evaluations of how these systems perform in real political and social contexts.

The findings reveal significant variation in chatbot performance across countries and languages. In some cases, systems provided nuanced, rights-based explanations aligned with international standards. In others, responses were incomplete, overly generic, or internally inconsistent, particularly when prompts touched on contentious political contexts or national-level human rights records. The study also highlights discrepancies between English-language outputs and responses generated in other languages, raising important questions about equitable access to reliable AI-generated information.

By offering comparative evidence from 15 countries, MJRC’s report seeks to inform regulators, developers, and civil society stakeholders about the democratic implications of free AI chatbot deployment.

See the study here.

See the AI study program page here.

Photo: Pexels