Arzo vs Kahubi

Arzo has no write-up on Specifiction yet; its figures are above.

Kahubi is an AI workspace for research writing, reading, and analysis using a researcher’s own papers, drafts, datasets, and transcripts. Its chat searches a library, answers from selected sources, and cites them. Manuscript tools support source-grounded drafting, PDF compilation, Word export, and LaTeX editing. A systematic review workflow searches OpenAlex and the web, screens papers against criteria, and creates PRISMA counts and a written review. Researchers can also create and analyze surveys, run CSV statistics including t-tests, ANOVA, regression, correlation, PCA, and Cronbach’s alpha, and analyze interviews transcribed in 99+ languages with speaker labels and timestamps. Shared projects offer viewer, commenter, and editor roles, with discussion channels and threaded comments. Integrations or supported formats include Zotero, Mendeley, GitHub, BibTeX, RIS, LaTeX, and Microsoft Word. The free plan includes 25 AI messages per month, two projects, 200 library papers, three drafts per project, and a one-time 30-minute transcription trial. Paid plans start at $29 a month. Kahubi says content is not used to train AI models and servers are in the European Union.