AI for Private Equity: How Investors Actually Use ChatGPT and Claude

An investor's week is heavy with a specific kind of work: turning documents other people produced into judgment other people will act on. A CIM lands and needs a first read before Thursday. The IC memo needs a draft. Twelve portfolio companies sent their monthly packages, and the partner meeting is Monday. The quarterly LP letter is due. A new thesis needs a market map before anyone commits analyst hours. Each of those tasks is reading-and-writing work wrapped around investment judgment, which is exactly the shape of work ChatGPT and Claude do well. Private equity AI, at the individual level, is not about replacing the model, the data room, or the partner's conviction. AI's seat in the week is narrower and, in practice, more valuable: first drafts of the language, fast reading of the documents, and a patient partner for stress-testing the thesis.

Each of these uses shares one property: every fact in them comes from the investor's own material, pasted into the chat. The AI tool never touches the data room, never queries a portfolio company's systems, and never produces a return figure anyone relies on without checking. That boundary—AI handles the language, your models and your people own the numbers—is what makes artificial intelligence in private equity safe enough to use and useful enough to matter.

Private equity carries risks most industries do not. The material on an investor's desk—deal terms, CIMs, LP commitments, unpublished portfolio-company results, and anything that could be material non-public information—is some of the most sensitive text in the economy. Three cautions matter twice as much here as anywhere else, and I cover all three with every investor I coach before the first real prompt.

First, confidentiality. The rule I give PE professionals is the rule I give every executive, doubled: treat anything you paste as shared with an outside vendor, because contractually it is. Consumer plans at OpenAI, Anthropic, and Google may use what you type to improve their models unless you change a setting; business-tier plans and APIs do not, by default—see OpenAI's data-use documentation and Anthropic's commercial terms. The practical floor for an investor: firm work runs through firm business-tier accounts with compliance sign-off, and deal terms, LP material, and anything resembling MNPI stay out of consumer accounts entirely—anonymized, as “a $200M industrial services platform,” when you want structural advice without the exposure. If your compliance manual does not address AI tools yet, that conversation comes before the first paste, not after.

Second, never trust the arithmetic. ChatGPT and Claude read figures well and compute them unreliably: give a language tool a long column or a multi-step return calculation and it will eventually get the math wrong, confidently, without flagging the error. The investor's rule is simple. AI may read your figures and write about your figures, but every IRR, MOIC, and covenant calculation gets recomputed in your own model before the draft goes anywhere. The words are the draft; the returns are yours.

Third, your compliance environment applies unchanged. If the firm answers to a compliance manual, an insider-trading policy, LP confidentiality clauses, or a regulator, then anything AI touches remains working material: drafted by a tool, reviewed by a person, sourced from documents the firm already holds. “The AI wrote it” is not an answer when compliance or an LP asks where a statement came from. Used properly, AI sits before the review, producing the first draft that your existing process then handles exactly as it handles any associate's work.

In most funds I coach into, someone on the deal team is already using AI—the associate summarizing the CIM, the analyst drafting the first pass of the memo. AI in private equity is not a rollout decision waiting to be made at the partner level; it is a quiet practice waiting to be brought into the open. The partner's move is the one-page set of norms: which tools and accounts are approved, what never goes in, and how AI-drafted material gets reviewed. A partner who has used the tools personally writes credible norms. A partner who has not writes fearful ones, and the team routes around them.

The compounding value sits in the fund's recurring documents: the IC memo format, the monthly portfolio summaries, the LP letter, the diligence question lists. A team that learns to brief an AI tool well—context first, material pasted, output verified—turns each recurring document into a saved prompt that improves every quarter. The professionals who learn the skill become more valuable to the firm, not less, and the skill itself is teachable in weeks.

Everything on this page can be done alone, this week, on a business-tier plan—and for many investors, self-teaching is the right start. The ceiling appears around month two: the generic workflows are running, and the open questions become specific to your fund. Which recurring document gets rebuilt first? What belongs in an AI-assisted IC memo, and what does not? How do you set the team's norms with your compliance manual and your LPs in mind? Those questions have no generic answers.

That ceiling is where a human teacher—not an AI coach bot, not another course—earns its keep. In a 1-on-1 session over video, we open your actual deal calendar—the sourcing cadence, the IC rhythm, the reporting cycle, the LP calendar—and build the prompts, the verification habits, and the team's ground rules around your fund and your risk tolerance. I work with investors across the US, remotely, on their real documents. This page is the map. Coaching is walking the route on your terrain.