A key part of being a great investor is to constantly be a student of the game and always learn.

Thankfully with AI we can now learn faster by breaking down new concepts or ideas.

This week, I listened to a podcast about Fiscal Dominance - a topic I was a total novice about.

The questions I had as an investor were:

What is Fiscal Dominance?
How does that impact the economy?
What is the probability it happens?
How do I make money off this? What do I invest in?

I'll show you how I leveraged AI prompting to go from novice to expert on this topic in just a few minutes.

Here's what we'll do:

Craft a detailed Deep Research prompt alongside ChatGPT
Paste the prompt in Deep Research boxes of ChatGPT, Gemini, Perplexity to compare results

(PS, you can use these same techniques to learn about any new topics you come across)

The Prompt

A prompt engineering technique I like to use for deep research is to break it down into two steps.

First, we craft the prompt together with the LLM with a Q&A to fill in missing gaps of detail.

Then, we take the AI written fleshed out prompt to paste into deep research.

This gives the best results in terms of hallucination minimization and quality of research.

Prompt #1 to Craft the Prompt

You are an expert prompt writer for ChatGPT Deep Research with a particular specialization for investing and macro research. For context, I am an equities investor and I would like to better understand (1) What Fiscal Dominance is (2) What the probability is of that happening in the United States in the near and mid term (3) The impact to the economy and different asset classes in this situation (4) Impact to my portfolio if I am primarily allocated to tech equities and digital assets (5) Which asset classes would perform well and underperform in a Fiscal Dominance scenario. .... before you craft the prompt ask me questions to clarify anything you may be uncertain of or if there are any gaps of knowledge we need to fill in together.

Prompt #2 for Deep Research

Deep Research Prompt — “Fiscal Dominance & Portfolio Implications”

Role & Voice

You are a senior macro-strategy analyst writing for an equities investor who holds • 30 % cash • 30 % large-cap U.S. tech equities • 30 % Bitcoin • 10 % broad-market index funds. Write in a direct, conversational yet professional style with clear action verbs and outcome-focused phrasing. Use Markdown headings, numbered and bulleted lists, and inline hyperlinks for sources. Keep sentences concise, avoid em dashes, and cite reputable data or academic work for every major claim.

Research Objectives

Define Fiscal Dominance – explain the concept, key mechanisms, and how it differs from Monetary Dominance.

Estimate Probabilities – assign probability bands for Fiscal Dominance emerging in the United States

Near term: next 6-12 months

Mid term: next 2-3 years Provide Base / Bull / Bear scenarios with justification.

Macroeconomic Impact – discuss likely effects on growth, inflation, rates, credit spreads, and the USD.

Asset-Class Impact – assess relative performance for Treasuries (nominal, TIPS) IG & HY credit Equities (by sector & factor), private equity Commodities (energy, industrial metals, gold) Real estate (public REITs, private)Volatility products (VIX, rate vol) FX (G-10 & EM) Digital assets (BTC, ETH, alt-coins, stablecoins)

Portfolio Diagnostics – evaluate the user’s current allocation under each scenario: Drawdowns / upside risk Correlations & beta shifts Stress points (liquidity, funding, regulation)

Winners & Losers – rank which asset classes (and key sub-sectors) should outperform or underperform if Fiscal Dominance takes hold; include brief rationale.

Historical Case Studies – summarize at least three precedents (e.g., post-WWII U.S., 1970s Italy, 2010s Japan), highlighting policy mix, market reaction, and lessons for today.

Actionable Takeaways – list practical steps the investor could take (hedges, position sizing, optionality plays, diversification ideas).

Key Risks & Unknowns – outline variables that could invalidate the analysis.

Formatting & Length

Target ~3,000 words (deep dive).Use H2 (##) for major sections, H3 (###) for subsections. Embed short bullet lists for clarity; avoid overly long paragraphs. Include a one-page executive summary up front.Provide numbered end-notes or inline bracketed citations tied to a “Sources” section.

Sources & Citations

Prioritize primary data and high-quality research: FRED, BEA, CBO, BIS, IMF, academic journals, Fed working papers, and well-regarded sell-side pieces. Cite every quantitative claim; hyper-link the source title rather than the raw URL. Flag any data with limited transparency.

Output Reminder (for the model)

Return only the finished report, starting with the executive summary. Do not restate these instructions.

The Result:

I tested the results of the Deep Research functions for ChatGPT, Gemini, and Perplexity.

I found the results from ChatGPT to be the best with Gemini and Perplexity coming in second and third respectively.

ChatGPT's results feels more 'investor facing' / slightly more nuance with the asset class suggestions whereas Gemini's was more academic - perhaps due to the fact that Gemini is trained with heavier emphasis on journals and papers.

Perplexity's results was solid for a quick and dirty analysis (fastest results of the 3) but lacked some depth vs the other two for this prompt.

Full Research Report Links: