Provenance Logbook
A running record of every AI interaction in your research workflow — prompts, responses, edits, and attribution level — that exports a verifiable provenance report you can show hiring committees, journals, or grant reviewers.
PhD candidates, postdocs, and researchers on the academic job market who use AI in their daily scientific work
- Browser and IDE plugin that captures prompts and responses tied to the manuscript, grant, or notebook you're working in
- Attribution tagging — generated, edited, fact-checked, idea-only — on every AI contribution
- Exportable provenance PDF formatted for academic job applications, journal disclosures, and grant reports
- 'Independence evidence' module that logs problems you solved without AI to balance the record
The viral chalk-talk post surfaces a real tension: search committees increasingly flag 'concerns about independent thinking' when AI use is suspected but undeclared — a verifiable, opt-in logbook turns that liability into a credential and forces a conversation about norms
Nature and Springer pieces document widespread ChatGPT use in academia and emerging AI-policy guidance for PhDs, but demand for an explicit provenance LOG is unproven — most researchers hide use rather than log it.ChatGPT turns two: how the AI chatbot has changed scientists' lives ↗Navigating AI Policies for PhD Students in 2025: A Doctoral Researcher ↗
The prompt-provenance dashboard space (PromptTrack Pro, ProvenanceHub, TraceAI Console, AuditPrompt Suite) is crowded but 100% enterprise/compliance-focused (HIPAA, GDPR, AI Act) — no tool targets academic job-market candidates or research workflows specifically.Review: Best AI Prompt Provenance Dashboard Tools for Traceability, Auditability, and Compliance ↗
Target users (PhD candidates, postdocs on the academic job market) are notoriously price-sensitive and conditioned to free tools; existing provenance platforms price for enterprise budgets, and there's no evidence of WTP for personal AI-audit reports in academia.ProvenanceAI: A Learning Platform for the AI-era ↗Google NotebookLM | AI Research Tool & Thinking Partner ↗
Real and growing institutional pressure (journal AI-disclosure mandates, funder transparency rules) supports multi-year relevance, but the use case could be absorbed by free IDE/extension tooling or mandated institutional logging within 2-3 years.Exploring the impact of AI on research practices: a postdigital ↗
Buildable as a browser extension + CLI wrapper that intercepts LLM API calls and exports signed PDF reports; technically straightforward (a few weeks for an MVP) but verification/trust layer adds non-trivial complexity.