Potentially Impactful Research: Unjournal AI-Assisted Prioritization Dashboard (~Prototype)

Key Takeaways

• Dashboard aggregates papers from NBER, arXiv, CEPR, SSRN, Semantic Scholar
• AI scores based on relevance, prominence, timing, and methodology
• Only abstracts and metadata are analyzed; full texts are excluded
• Unjournal plans a centaur model blending AI suggestions with human review

Pulse Analysis

The rise of large‑language models has sparked a wave of automation in knowledge‑intensive fields, and research prioritization is a natural next frontier. By feeding paper titles, abstracts, and metadata into GPT‑5.4‑class models, Unjournal’s prototype can quickly surface studies that align with its strategic criteria. This approach mirrors broader industry moves where AI assists analysts in sifting through ever‑growing scholarly output, reducing the manual labor traditionally required to keep policy‑relevant research pipelines full.

Unjournal’s dashboard pulls from a curated list of high‑quality repositories—NBER, arXiv’s economics section, CEPR, SSRN, Semantic Scholar, the EA Forum, OpenAlex, and Anthropic Research—then evaluates each entry against four dimensions: decision relevance, prominence, timing value, and methodological potential. While the current version only sees abstracts, the scoring system provides a transparent, quantifiable signal of expected review value, not intrinsic research quality. The prototype’s statistics page breaks down source contributions and score distributions, offering early insights into which domains (e.g., AI governance, catastrophic risk forecasting, biosecurity) are generating the most actionable ideas.

The broader implication is a shift toward hybrid "centaur" workflows, where AI-generated rankings feed human experts who can validate, adjust, or reject recommendations. Such a feedback loop promises to improve calibration over time, making the system more reliable for commissioning independent reviews or guiding grant allocations. If refined, this model could become a staple for think tanks, NGOs, and government agencies seeking to allocate limited resources toward research with the highest policy impact, ultimately accelerating evidence‑based decision‑making in a fast‑moving world.