They’re not just automating tasks. They’re building an unfair advantage — and you can too, starting today.

So, how to use AI in finance? The top 1% never ask AI to think of them; they ask it to pressure-test the thinking they’ve already done. Before you type a prompt, ask yourself: ‘What is my actual thesis here?’ The magic doesn’t happen when AI solves the problem from scratch. It happens when you bring a complex problem to the table and use AI to find the blind spots you couldn’t see alone.

They’re asking ChatGPT to summarise an earnings report. They’re using it to auto-fill spreadsheets. They feel productive. They share screenshots on LinkedIn.

Meanwhile, the actual top 1% — hedge fund analysts, family office managers, CFOs at fast-scaling startups — are using AI in a way that changes the quality of their decisions, not just the speed of their work.

That gap matters more than people realise. Because in finance, the edge isn’t information anymore. Information is free. The edge is what you do with it, and how fast.

“The analyst who masters AI won’t replace every analyst. But every analyst who doesn’t will be replaced by one who does.” — not a cliché, just math.

Let me walk you through exactly how the sharpest financial minds are using AI right now — not in theory, in practice.

1. They use AI as a thinking partner, not a search engine

The biggest mistake? Asking AI for answers. The top 1% ask AI to pressure-test their thinking.

Before making an investment call, they’ll drop their entire thesis into a model and ask: “What’s the strongest counterargument to this? What am I probably wrong about? What would have to be true for this to fail?”

That’s not Googling. That’s having a senior partner review your memo at 2 am, for free, without judgment.

Try this! (How to use AI in finance)

The Pre-Mortem Prompt

Paste your investment thesis or financial decision. Then ask: “Assume this was a catastrophic failure. Walk me through the 5 most likely reasons why it failed, in order of probability.” Then defend against each one.

2. They turn raw data into a readable signal — instantly

A 10-K filing is 200 pages. An earnings transcript runs an hour. A macro report buries the insight in paragraph 14.

Top analysts aren’t reading all of it anymore. They’re using AI to extract only what changes the model — the one number, the one management comment, the one footnote that shifts the outlook.

They’re uploading PDFs, pasting transcripts, and asking hyper-specific questions: “What did management say about margin guidance compared to last quarter? Flag any language that sounds more cautious than before.”

The AI doesn’t replace the analysis. It removes the noise so the analyst can go deeper, faster.

Try this! (How to use AI in finance)

The Signal Extractor

Upload any earnings call transcript. Ask: “List every forward-looking statement management made. Then flag the ones that are more hedged or cautious than statements from the previous quarter’s transcript.” Paste both for comparison.

3. They build scenario models at conversation speed

Traditional scenario analysis takes a junior analyst a day. You rebuild the model for base, bull, bear. You change 12 assumptions. You format it. You present it.

Now? The best CFOs and investors are doing scenario analysis in a live conversation. They describe the business, the key levers, the uncertainty, and ask AI to walk through how changing one variable flows through to the bottom line.

It’s not replacing Excel. It’s replacing the time between question and insight.

Try this! (How to use AI in finance)

The Lever Conversation

Describe your business or portfolio position in plain language. Tell AI the 3 variables you’re most uncertain about. Ask: “If variable X moves 20% worse than expected, what downstream effects should I be modeling? What would I need to hedge?” Let it build the mental model with you.

4. They use AI to simulate the other side of the table

Negotiating a term sheet? Pitching a board? Going into an LP meeting?

The smartest people are using AI to roleplay the hardest version of those conversations — before they walk in the room.

“You are a sceptical LP who’s seen three similar fund pitches fail. Ask me the 10 hardest questions you’d want answered before committing capital. Don’t hold back.”

That kind of prep used to require a trusted mentor, a senior advisor, or expensive consultants. Now it happens at midnight before the meeting.

Try this! (How to use AI in finance)

The Adversarial Roleplay

Before any high-stakes financial conversation, ask AI to play the other party at their most skeptical. Give it context about who they are, what they care about, and what their objections usually are. Run the conversation. Refine your answers.

5. They build systems, not habits

This is the one that separates the serious operators from the curious experimenters.

Using AI once, well, is a good session. Building a repeatable process — a set of prompts, workflows, and checks you run every week — is a structural advantage.

The top 1% have essentially built their own private research workflows. A weekly macro brief prompt. A portfolio review template. A deal evaluation checklist that runs through AI before anything hits a spreadsheet. Not because they’re tech people — because they understand that consistency compounds.

Try this! (How to use AI in finance)

The Weekly Intelligence Loop

Write one standing prompt you’ll run every Monday: “Here are 5 macro data points from last week. Based on these, what changed in the outlook for [your sector/portfolio focus]? What should I be watching more closely this week?” Refine it over time. Make it yours.

One thing the top 1% never do

They never outsource their judgment to AI.

They use it to sharpen their judgment — to see angles they’d miss, to stress-test positions they’d otherwise be too close to, to process volume they’d never get through manually.

But the final call? The risk they’re willing to take? The conviction they’re building over time? That’s still human. It has to be. Because the market doesn’t care about your AI prompt. It only cares if you were right.

AI just gives you a better shot at being right, more often, with fewer blind spots.

That’s the actual edge. And it’s available to anyone willing to use it seriously.

The best time to start using AI this way was a year ago.
The second-best time is today’s first prompt.

For more blogs like this, check out tryFluentra