Compare the capabilities of OpenAI Deep Research and Gemini Deep Research in generating comprehensive reports on private SaaS companies in this 37-minute video analysis. Explore how each AI system handles challenges like data scarcity, source credibility assessment, industry jargon comprehension, and predictive analysis. Through detailed examples and systematic evaluation, discover OpenAI's superior performance in report structure, market analysis, financial assessment, and strategic forecasting, while noting Gemini's strength in recent fact incorporation. Learn how these AI tools approach complex business research tasks, including market share analysis, SaaS metrics evaluation, and IPO predictions, with practical implications for business intelligence and market research workflows.
Comparing OpenAI and Gemini Deep Research for Private Company Analysis
Overview
Syllabus
0:00 Introduction
1:06 Data scarcity with private companies
2:10 The agent needs to “fill in the blanks”
2:36 The agent needs to weigh the credibility of sources
4:02 Contextual understanding of industry acronyms and jargon
4:43 Can deep research agents make predictions with sparse information?
6:24 My prompt for report generation on a niche market
10:52 Me as a human judge of the output
11:57 LLMs as judges prompt - OpenAI, Claude, Gemini judge the reports
13:30 Begin walkthrough of the complete deep research report
14:14 Structure of each report OpenAI is better
15:10 Length and depth of each report OpenAI is better
16:25 Market overview and key player identification OpenAI is better
19:00 Market share analysis OpenAI is better
20:41 Financials and Valuation OpenAI is better
23:52 Determining SaaS metrics and public comps OpenAI is better
26:45 Strategic outlook and predictions Gemini’s response is boring
27:51 OpenAI makes bold, specific predictions, IPO outlook
29:48 OpenAI gives 2025 predictions on market share, M&A
31:02 Gemini captures 2 recent facts that OpenAI missed
32:25 Was this caused by a knowledge cutoff?
33:10 Getting the best of both worlds, going deeper?
34:17 LLM judges - what do they think about the reports?
36:18 Wrapping up, RIP Substackers?
Taught by
Part Time Larry
