Quarterly AI Update for Valuators 

Welcome to a new feature in ValuSource Insider! This month, we’re running the first of what will be regular updates on the latest developments in AI that valuators need to know about, including useful advice on how you can begin — or continue — to introduce ground-breaking tools into your valuation practice.

In this update, we’ll share some highlights from a recent discussion with Nainesh Shah, Managing Partner – Valuation Advisory with Complete Advisors in New York City and a prominent expert specializing in integrating AI into business valuation and financial forensics. The conversation touched on a several notable areas:

1. How valuators can learn AI

Most valuators, whether they conduct dozens of valuations per year or just a handful, understand that AI has uses for their practice, but they don’t necessarily understand how they can apply it. Nainesh says you need to invest the time and effort to learn how it works if you want to avoid costly missteps.

He advises valuators of all levels to attend intensive courses, such as NACVA’s AI Boot Camps. These camps run eight hours a day for two days for the in-person session or four days, four hours a day for the virtual session, and are available throughout the year. You can attend an in-person or virtual course, but Nainesh recommends the in-person sessions when possible. Being face-to-face makes hands-on learning easier.

2. The importance of verification

GenAI continues to produce hallucinations in its results — a significant risk for valuators who use it for professional purposes. In one webinar Nainesh led, he provided the participants with a prompt that they all used while uploading the same PDF. “Almost everybody got a different answer,” he says.

The lesson? Every quantifiable result must be double-checked before you use it, particularly for valuation purposes.

3. Areas where the efficiencies outweigh the burden of verification

For all the caveats about hallucinations and the need for verification, Nainesh says he’s a heavy user of AI — just not to build reports. He identified four examples of use cases for valuators:

• Summarizing industry reports. For instance, he’ll upload a PDF of an industry report and ask Copilot Pro to summarize it in three paragraphs. To verify, he instructs the tool to provide footnotes showing where in the report it obtained various pieces of information.
• Converting PDFs to Excel. Valuators need to do this often, and AI is effective for streamlining conversion. An attendee at Nainesh’s webinar converted a large income statement into Excel, where he was then able to create graphs, etc.
• Writing new types of reports. If a valuator has never written a certain kind of report, AI can prove helpful. You could, for example, ask it to “provide a table of contents for an X report.”
• Handling smaller parts of the valuation process. Valuators can confidently rely on AI for smaller tasks, such as drafting a DLOM or company-specific section of a report. As the technology expands, Nainesh predicts, AI will be able to combine all of the smaller components into a comprehensive report — but he cautions that it hasn’t yet evolved to that point.

4. How you can teach AI your writing style

If you send someone an email written by AI, Nainesh says, most people can sense AI’s role because the text doesn’t sound like you (or, sometimes, like any human). You can avoid that with some AI tools, though. For example, Copilot now has a new “draft instructions” setting that those with a Microsoft 365 Copilot or Copilot Pro license can use to make AI write more like you. Every AI tool has a similar option. You might instruct it to “keep emails short and direct, with bullet points where they make sense. Use a friendly tone and sign emails with my first name.” You can add a lot of detail or keep it simple.

5. Using AI to improve your prompts

Not happy with the results you get from AI tools after you enter a prompt? Often, the problem lies in the prompt itself. The good news is that Nainesh has written a “prompt of a prompt” that valuators can use to optimize their prompts. You can find it and a wealth of other valuable tools in NACVA’s AI Data University. In general, Nainesh suggests using AI to help write prompts. For example, you can enter a draft prompt, ask it to “make it better,” and then submit the revised version.

Stay tuned! We’ll share our next quarterly update in the July 2026 issue of ValuSource Insider. If there’s one thing that’s undeniably true about AI, its’sthat new developments are coming fast and furious — and you can’t afford to fall behind.