• Fiscal.ai
  • Posts
  • 🗞️ 4 Best Ways to Use AI in Investing

🗞️ 4 Best Ways to Use AI in Investing

Models, research reports, recurring scans, and more.

Happy Sunday! đź‘‹

Over the last couple months, we’ve had several investors ask us for tips on incorporating AI into their research process. So we’ve put together this list of 4 simple use cases to help with diligence and idea generation.

Let’s dive in!

4 Best Ways to Use AI in Investing

There has never been a better time to be a stock research analyst.

With AI and the right financial data, you can now eliminate 95%+ of the grunt work (manual data entry, building models from scratch, digging through filings, etc).

By automating that busywork, investors can spend more time refining their assumptions to drive better investing decisions.

While the applications for AI in investing are endless, we’ve identified 4 effective ways to put it to work today.

  1. Research Reports

If you want to go from zero to one on a new idea, an AI Research Report is a great place to start.

The Fiscal.ai research template includes the history, business model, competitive landscape, moat, financials, capital allocation, investor narrative, and much more.

Here is the full research report.

With context from text-rich sources like news, fund letters, and conference calls, the Fiscal.ai research report is able to deliver real qualitative and opinionated substance to provide a better understanding of the current setup.

If you already have the Fiscal.ai MCP connected, just type “/fiscal-investment-research” then input the ticker you want to research.

If you don’t have the MCP, here’s a full prompt you can copy and paste to try for yourself.

  1. Building Financial Models

Investors can now build a complete, 3-statement financial model in a fraction of the time that it would have taken them 5 years ago.

No more building models from scratch and manually parsing filings to find the numbers you need.

Here’s a full model for MercadoLibre that the Fiscal.ai research team built this week.

Quick note: Financial modeling (more so than other queries) requires a data connector to get a great output. The financials, segments, KPIs, estimates, conference calls for context, and filing pages for sources are impossible to procure without a structured database providing it. The Fiscal.ai MCP comes with a pre-built modeling skill (and 23 other ready-made prompts) that includes all of this.

  1. Creating Visuals

Data visuals go a long way, especially in investing.

Whether you’re building a bar chart, scatter plot, or market heatmap, turning raw financial data into clean visuals can help spot patterns and support an investment thesis.

We created the graphic above with the following prompt:

“Build me a heatmap of all the stocks in the S&P 500 currently in more than a 50% drawdown from their 10-year highs.”

  1. Recurring Screens/Trackers

If you’re looking to generate new ideas, setting up recurring screens/trackers can help you catch new ideas before other investors.

Every morning the Fiscal.ai research team receives a series of notifications from AI with the following:

  • All Insider Buys over the last 24 hours

  • All Politician Trades over the last 24 hours

  • New Fund Letters published (with associated buys and sells)

  • Quarterly Earnings Summaries (Beats/Misses, Guidance Revisions, Conference Call Quotes, etc.)

Here’s a prompt you can copy and paste to generate a daily earnings recap:

Every run, find the 5 largest companies by market cap that reported earnings after the market close on the previous trading day, using the Fiscal.ai connector for the earnings calendar, results, estimates, guidance and call transcripts (fall back to company press releases and reputable news only for anything Fiscal.ai doesn't carry, and say so). For each company, cover: reported revenue and EPS against consensus, stated as beat/miss/in-line with the dollar and percent difference, plus 1–2 key segment or KPI results; any change to guidance (raised, lowered, maintained, initiated or withdrawn), with prior and new ranges compared to consensus; 2–3 notable quotes from the conference call, attributed to the speaker, focused on demand, margins, outlook or capital allocation; and the after-hours or premarket stock move if it's available. Don't estimate or invent any figure. If a call hasn't happened yet or a number isn't available, label it as pending.

Publish the result as a single artifact titled "After-Hours Earnings Recap – [date]". Put a 3–4 sentence summary of the overall tone at the top, then one card per company in market-cap order. Each card should show the company logo, a beats/misses scorecard, guidance changes and the quotes, with a source link for every figure. Keep it scannable with about 150–200 words per company. If fewer than 5 companies reported, cover the ones that did and say how many. If none did, publish a one-line note instead of an empty recap.

Recurring scans like these can keep investors up-to-date on new market developments, and help inspire new ideas for further investigation.

That’s all for this week.

If you have any other ways that you like to leverage AI in your investing process, let us know!

If you have any other questions about Fiscal.ai, simply reply to this email.