In the video, Dave showcases his AI-powered research stack—combining tools like GPT-3.5 Pro, Deep Research, NotebookLM, and others—to produce detailed, publicly accessible research papers on post labor economics, efficiently managing and synthesizing vast information. He emphasizes the use of AI for deep analysis, diverse ideological exploration, and continuous refinement, aiming ultimately to create a comprehensive book while encouraging open collaboration through Creative Commons licensing.
In this video, Dave shares his comprehensive AI-powered research stack that he uses to conduct deep and high-quality research on post labor economics. He begins by showcasing the final output of his research process: a publicly accessible GitHub repository filled with purpose-built research papers in PDF format, all generated and organized using various AI tools. These papers are compiled into a website via GitHub Pages, making it easy to browse through his extensive research collection. Dave emphasizes that each paper is tailored to specific topics or subtopics, ensuring focused and detailed exploration of the subject matter.
Dave highlights the powerful combination of OpenAI’s GPT-3.5 Pro and Deep Research as the core of his research workflow. GPT-3.5 Pro allows for longer, more thoughtful responses akin to expert-level essays, which he uses to explore complex geopolitical and economic questions. Deep Research then validates, refines, and synthesizes these insights, helping Dave produce comprehensive reports. He demonstrates this with an example where he investigates global geopolitical tensions and their implications for post labor economics, culminating in a detailed 30-page report that ties together economics, conflict, and automation.
To manage and explore his large repository of PDFs, Dave uses Google’s NotebookLM, which excels at handling vast amounts of data with a large context window and offers a visual mind map feature. This tool helps him navigate and query his research corpus efficiently, making it especially useful for thesis or dissertation work. NotebookLM can generate summaries and contextual insights from the uploaded documents, providing a deep research experience focused on his own curated sources. Dave notes that while NotebookLM’s model isn’t the smartest yet, its ability to handle large datasets and visualize connections is invaluable.
Dave also incorporates a feedback loop using AI tools like Claude, Grok, Gemini, and ChatGPT with internet search to gather contemporary reactions and critiques of post labor economics. This approach helps him identify areas of consensus and disagreement within the academic and social discourse, enabling him to refine his work continuously. He uses these insights to conduct rapid literature reviews and synthesize viewpoints from prominent economists, highlighting key agreements such as the structural threat of automation to wage labor, the need for broad-based capital ownership, and the importance of active policy steering during the economic transition.
Finally, Dave explains how he uses his AI stack to explore different ideological perspectives, such as a Marxist reading of post labor economics, revealing both agreements on diagnosis and divergences on solutions. He stresses that his research materials are openly licensed under Creative Commons Zero, encouraging others to use and build upon his work. While the current output consists of research papers and reports, Dave’s ultimate goal is to produce a comprehensive book on post labor economics. His AI-driven research stack exemplifies how modern tools can accelerate and deepen academic research, making complex topics more accessible and well-documented.