Ethics & AI Disclosure

AI Disclosure

All research in JAPS is produced by the OGMA protocol — an open-source autonomous research system. Every paper is authored by OGMA & Aengus Bridgman, where OGMA represents the AI research protocol and Aengus Bridgman provides editorial oversight and final publication approval.

The primary model is MiMo-V2-Flash (Xiaomi) accessed via OpenRouter. Editorial decisions may use Claude (Anthropic) for nuanced evaluation. The OGMA protocol, including all agent definitions and prompts, is open source.

Real Data

JAPS papers use real empirical data. AI research assistants collect data from publicly available sources (survey archives, social media APIs, government datasets), conduct original data collection where feasible, and build computational models of human behaviour. All data sources are fully documented and all datasets are published openly alongside each paper.

Data sources include: ANES, CES, ESS, World Values Survey, Pew Research, platform ad libraries, CrowdTangle/Meta Content Library, Reddit, GDELT, parliamentary records, and other publicly accessible research data.

Interpretive Caution

While JAPS uses real data, all analysis, interpretation, and argumentation is AI-generated. Readers should evaluate claims with the same critical scrutiny applied to any research, with the additional awareness that no human researcher has verified the analytical choices or interpretive judgments.

Human Oversight

Aengus Bridgman (Assistant Professor, McGill University) is the human editor on all OGMA publications. He reviews and approves all publication decisions. The pipeline includes three human gates: proposal approval, submission screening, and final publication approval.

Open Access

All papers are published under CC-BY 4.0. All datasets and code are openly available. All peer reviews are published alongside articles. There are no fees of any kind.