AI Prompts to Evaluate Thematic Investing on Wall Street

Welcome to AI in Investment Research & Finance, your weekly guide to how financial professionals apply AI to investment analysis and portfolio strategy.

We’ve previously explored how fund analysts and strategists use large language models (LLMs) to transform raw data into sharper insights. We’ve also covered value strategies, macro signals, and exchange-traded fund (ETF) positioning across funds including MOAT, SPLV, JEPI, GLD, BITO, EDOG, AIQ, QGRW, OMAH, QAI, FLOT and IVV, plus thematic comparisons like Gold vs Bitcoin and SLV vs SIL.

This week, we focus further on thematic investing, not to chase trends, but to demonstrate how well-crafted prompts can help uncover which themes are truly supported by data, fundamentals and consistent signals rather than hype.

Table of Contents

• This Week's Focus: AI Prompts for Thematic Investing

  • Message From Our Partner: Delve

    • AI Agents in Action: Thematic Research Gets an Upgrade

    • AI Career Moves: Exciting AI Jobs This Week

Thematic investing has exploded in popularity, with funds targeting numerous topics, ranging from from AI and clean energy to aging populations. But is it delivering real value or just narrative-driven speculation?

The numbers tell a sobering story: according to a recent Financial Times analysis of Morningstar data, only 20% of thematic ETFs have outperformed their benchmarks over the past five years. On average, these funds lagged broad market indices by 8.5 percentage points, a striking underperformance that reveals a critical flaw in how themes are evaluated and timed.

The pattern is predictable: many funds launch after themes gain momentum, when valuations are already inflated. Once initial enthusiasm fades, returns suffer and a number of those funds close. More than 130 thematic ETFs have shut down in the past 18 months, exceeding new launches.

Understandably, this doesn't mean investment themes don't matter. Bit it highlights that investors need better tools to separate themes with staying power from those simply priced for perfection. This is where structured AI prompting can transform thematic analysis from trend-chasing into rigorous fundamental research.

As thematic product launches accelerate, the competitive edge doesn't come from speed. It comes from asking better questions and prompting for answers that go deeper than surface narratives.

Prompt Examples: Building Smarter Thematic Screens

The prompts below support structured, AI-enhanced analysis across earnings calls, regulatory filings, fund flows and sector trends. As you use these prompt, assign the LLM a clear analyst role (e.g., “equity analyst” or “portfolio strategist”), request citations when possible and specify the desired output format, such as a bullet-point summary, chart or investment memo.

📌 Prompt 1: Backtest a Theme Across Cycles

Evaluate the long-term performance of clean energy ETFs [ETF Names] versus the S&P 500 over the past five years. Identify how the theme performed across different macro environments, including interest rate cycles and oil price shocks. Highlight the main drivers and enduring trends for each period.

📌 Prompt 2: Validate Theme Fundamentals

Identify the top five S&P 500 companies most fundamentally aligned with the “aging population” theme. For each company, assess revenue growth, return on invested capital (ROIC) and free cash flow trends over the last five years. In your analysis, highlight patterns that illustrate how these companies’ fundamentals have benefited from demographic shifts toward an older population in the US.

📌 Prompt 3: Spot Theme Overcrowding

Analyze recent fund flows into AI-themed ETFs [ETF Names] over the past 12 months to assess where capital has concentrated. Compare their current valuation metrics—such as price-to-earnings (P/E) and price-to-sales (P/S) ratios—against their own long-term sector averages and those of broader technology benchmarks. Identify signs of potential overcrowding or speculative excess, such as valuation premiums, sharp swing in flows or deviation from historical growth fundamentals. 

📌 Prompt 4: Monitor Theme Stickiness

Identify which investment themes that surged in popularity during 2020–2021, such as clean energy, cybersecurity, digital transformation, and longevity, are still represented by profitable ETFs or companies based in the US. For the durable themes, analyze the common characteristics that have supported their sustained profitability, such as structural growth drivers, resilient end-market demand, adaptable business models or proven paths to monetization. Summarize what sets these persistent winners apart from themes that faded, referencing turnover in ETF product ranges, shifts in investor flows or the ability of underlying companies to deliver consistent earnings growth despite changing macro conditions.

AI Agents in Action: Thematic Research Gets an Upgrade

Static prompts are powerful, but what if you could automate these analyses and run them continuously? OpenAI recently introduced ChatGPT Agents, capable of executing multi-step tasks, querying data, reasoning across documents, and automating research workflows, all with persistent memory.

Moreover, Anthropic recently launched Claude for Financial Services, specifically designed for investment research with native integrations to data providers like Snowflake, S&P Global and Morningstar. Bloomberg Terminal has integrated agentic AI capabilities, while AWS's Bedrock offers multi-agent systems that analyze financial news, evaluate stock performance, and optimize portfolio allocations through natural language interfaces.

The key difference: While ChatGPT Agents offer flexibility and customization for any workflow, these purpose-built financial agents come with deeper data integrations and industry-specific capabilities out of the box. Think plug-and-play versus build-your-own.

These agents share a common capability: executing multi-step tasks, querying data, reasoning across documents and automating research workflows, all with persistent memory.

Give one of these agents an investment theme and a mandate and you get something close to a thematic research assistant:

• Scans earnings calls and SEC filings for updates related to a theme (e.g., "industrial automation" or "precision medicine");

• Flags companies that increase or reduce their exposure;

• Backtests ETF baskets or screens for fundamentals across custom universes;

• Tracks new thematic ETF launches and structural shifts.

📌 Example Agent Brief: Longevity Economy Tracker (Potential weekly tasks to be assigned)

1. Detect and list all newly US-listed public companies or recently launched US-based ETFs explicitly focused on the longevity economy. Specify relevant listing dates, sectors and a brief rationale for inclusion under the longevity theme. Highlight any unique business models or innovations addressing the needs of an aging population.

2. Aggregate and summarize the latest earnings call references from S&P 500 companies (all US-listed) discussing key longevity-linked themes such as "aging population," "senior care" or "retirement." Extract and synthesize notable management commentary, explicit references to demographic trends and any announced strategic initiatives. Highlight direct business impacts or long-term strategic pivots driven by demographic change.

3. Benchmark the current price/earnings (P/E), price/sales (P/S) and EV/EBITDA ratios of the top 10 US-listed companies within the S&P 500 index that are most exposed to the longevity theme against their own five-year historical averages. Flag companies with any ratio deviating by more than 20% (premium or discount) from their historic mean. Add sector peer comparisons and, for flagged names, summarize likely drivers behind the divergence—referencing recent company announcements, results or macro developments. Present findings in a structured table with deviation percentages and key insights.

4. Monitor and alert if any US-listed ETF or mutual fund shifts its portfolio allocation to the longevity economy theme by more than 5% (either up or down) during the past quarter. Specify the fund, magnitude of change and whether this represents a trend or a single major allocation event. Where possible, indicate corresponding fund rationale from public communications.

5. Track and summarize notable US regulatory updates, patent filings, or federal policy changes directly affecting sectors within the longevity economy (e.g., healthcare, senior housing, retirement services). Highlight events most likely to impact theme fundamentals within the next five years and indicate the affected sub-sectors and potential strategic implications for investors.

This kind of agent doesn’t just save time, but potentially ireshapes how professionals link structural change to portfolio construction. The global AI agents market in financial services is projected to grow from $490.2 million in 2024 to $4.49 billion by 2030, signaling that this transformation is just beginning.

Key Takeaway

The most successful thematic investing strategies don't predict the future, they systematically analyze how fundamental changes are already reshaping markets. AI-powered prompting gives you the tools to make that distinction with precision.

AI Career Moves: Exciting AI Jobs This Week

Looking for your next opportunity in AI? Explore these standout roles across industries and locations:  

📩 Feel free to share this list with anyone looking for AI opportunities!

Stay tuned for next week’s edition, where we’ll explore new AI prompts for deeper sector analysis. To ensure our next newsletter lands in your inbox, please add our email address to your contacts: [email protected]