The academic research workflow platform

From idea to manuscript: a complete research lifecycle

A 32-step AI-augmented framework across 5 phases. Citations only from academic database searches. Human judgment at every decision point.

The goal is to amplify the researcher.

32 steps across 5 phases

From project genesis through manuscript development. Every stage guided by AI, every decision made by the researcher.

Iterative Improvement Cycle

Plan, Create, Critique, Refine. The cycle runs at every stage, ensuring continuous improvement until publication-ready.

Citations only from databases

Every citation traces to OpenAlex, Semantic Scholar, or CrossRef. The architecture makes hallucinated citations impossible.

Core Philosophy

Three pillars of the framework

Research is not linear. Neither is this framework. These three pillars guide every capability.

Pillar 1

AI-Guided Socratic Dialogue

AI does not answer for the researcher. It challenges the researcher's thinking. Through targeted questioning, it surfaces hidden assumptions, exposes gaps, and pressure-tests alignment before the researcher invests time building on a flawed foundation.

Pillar 2

Continuous Iterative Improvement

Every output is a draft. The framework supports repeated cycles of creation, critique, and enhancement. Work is never locked. It evolves as understanding deepens, literature reveals new insights, and methods sharpen.

Pillar 3

Human-in-the-Loop Decision Making

AI augments, humans decide. The researcher reviews, approves, revises, and directs at every stage. No AI output advances without human judgment. The researcher remains the author, the thinker, and the authority.

Why AI Augmentation?

AI processes literature at scale, identifies patterns humans miss, maintains consistency across complex documents, and never tires of asking "why?", freeing the researcher to focus on insight, judgment, and original thinking.

Why Human-in-the-Loop?

Research demands judgment that AI cannot provide: determining what matters, what is novel, what is ethical, and what contributes to the field. Every AI output is a recommendation, never a decision. The researcher's expertise, context, and values remain the driving force at every stage.

Peer-reviewed output produced through the platform

Western Economic Association International
American Marketing Association

Two peer-reviewed conference proceedings: WEAI 2026 and AMA Summer Conference 2026. Eight months of founder testing on real research.

32
Steps
5
Phases
8
Socratic Processes
20
Manuscript Sections
5
Reports

The Framework

From idea to manuscript

32 steps across 5 phases. AI-guided at every stage. Human decision-making throughout.

Phase ISteps 1-9
Operational

AI-Guided Socratic Dialogue: Foundation & Conceptualization

From initial idea to search-ready. AI challenges assumptions, pressure-tests research questions, and validates alignment through iterative Socratic dialogue.

3 Socratic Processes
Phase IISteps 10-14
Operational

Automated Agentic Literature Search & Evaluation

Fully autonomous pipeline. Multiple AI engines execute search, enrich metadata, screen for relevance and novelty threats, and produce deep reviews with no human intervention.

Fully Automated
Phase IIISteps 15-17
Operational

Human Literature Evaluation

Researcher validates AI results, adjudicates edge cases, records evaluations, and conducts targeted gap-filling searches that loop back through the agentic pipeline.

5 Reports
Phase IVSteps 18-32
Operational

AI-Guided Research Architecture

Theory foundation through methods design. Each workflow follows Socratic planning, creation, AI enhancement, and critique. Includes Libby Diagram and full-project evaluations.

5 Socratic Processes
Phase VSteps Per-Section
Pilot

AI-Guided Manuscript Development

Every section follows a 5-step workflow: Socratic Planning, Outline, Draft, Critique, Enhance. AI guides, researcher approves. Iterative until publication-ready.

Worldwide Collaborative Environment

One platform. Worldwide collaboration. Transformative research.

The greatest challenges of our time are not isolated problems: lack of economic opportunity, social disparities, environmental degradation, poor health and wellbeing, educational gaps, and barriers to entrepreneurship. They are interconnected crises that demand interconnected solutions.

Interconnected SolutionsWorldwide TeamsCross-DisciplinaryCascading ImpactShared AuthorshipReal-Time Access

The Research Stack

Where Orbis Scientia fits

Not a search tool. Not a reference manager. Not a general-purpose assistant. A full research workflow platform.

Reference Managers
← Orbis Scientia exports here
ZoteroPaperpileEndNoteMendeley

Downstream organizing layer for citation management

Orbis Scientia
32-Step Framework
Full research lifecycle from idea to manuscript
Citation Provenance
Verified academic database sources only
Audit Trail
Every action timestamped and attributed
AI Engines
ChatGPTClaudeGemini100+ models orchestrated

Foundation layer, orchestrated through OrbisFramework infrastructure

AI Search Tools
Elicit
Consensus
Research Rabbit
Scite
Undermind

Orbis Scientia replaces these.

50–100 search terms generate 1,000–4,000+ articles. Automated screening reduces this to 20–40 articles for human review, each with DOI links. One complete provenance ledger. No manual tracking. No spreadsheets.

The research stack shows three layers: At the bottom, AI engines like ChatGPT, Claude, and Gemini with 100+ models orchestrated. In the middle, Orbis Scientia provides the 32-step framework, citation provenance, and audit trail. At the top, reference managers like Zotero, Paperpile, EndNote, and Mendeley handle citation organization. AI search tools like Elicit, Consensus, and others provide partial scope overlap for ad hoc search work.

Research Program

The FT50 Research Program

Four studies, 50,000+ articles. The platform is the instrument, the program is the science, the publications are the evidence.

Study 1

Pilot

Meta-Analysis of FT50 Scholarship

The November 2022 inflection point as a structural break.

Comprehensive meta-analysis of FT50 journal publications examining the structural break in research patterns following the November 2022 inflection point.

  • Theory library seeding
  • Constructs and variables library
  • Methodological pattern identification

Study 2

Ready for Pilot

Industry Applicability Review and Executive Summarization

Three value paths for academic research translation.

Systematic review and executive summarization of FT50 research for industry applicability.

  • Executive access to FT50 insights
  • MBA and undergraduate curriculum grounding
  • Practitioner translation

Study 3

In Development

Computational Research Gap Identification

Scaled research gap analysis for university research programs.

Computational identification of research gaps using OpenAlex and Semantic Scholar cross-reference, scaled to coordinated university research programs.

  • Institutional research coordination
  • Doctoral program gap mapping
  • Cross-disciplinary opportunity identification

Study 4

Research

AI-Augmented Research Design

The platform as research instrument.

Methodological study of AI-augmented research design using the platform itself as the research instrument. Pre-registration noted when in place.

  • Platform validation
  • Methodology contribution
  • Reproducibility demonstration

Roadmap

A research methodology agenda that ships as software

Five phases of development. Commitments backed by working capability, not aspirations.

Phase 1Operational

Foundation

Phase 2Pilot

Meta-Analysis

Phase 3In Development

Quantitative

Phase 4Planned

Intelligence

Phase 5Research

Qualitative

Built by a researcher, for researchers

Forty years in high-stakes environments.
Doctoral candidate at Daniels College of Business, University of Denver.
Eight months of pre-release testing on real research before public release.

Full story at bradleywpetersen.com →

Ready to amplify your research?

Schedule a strategic conversation to explore how Orbis Scientia fits your research workflow.