Here are three ways I use AI for cross-border due diligence—and two things it still can’t do

Last month, our team reviewed an Indian Ayurvedic medicine company for a GCC-based investor. This was a new area for me—traditional medicine looking to reach modern markets. Ten years ago, this would have taken weeks with consultants and lots of costly research before even deciding if it was worth it. This time, we finished the initial screening in just a few days. Here’s what went well, what didn’t, and the exact prompts I used, so you can try them in your own cross-border projects.

What AI Does Well: Three Useful Examples

• Mapping the Regulatory Landscape: India’s Ministry of AYUSH oversees the Ayurvedic sector, which is a set of rules most finance professionals outside India may not know. I needed to quickly learn about licence requirements, export certifications, and compliance standards.

Prompt template:

Give a full outline of the rules and regulations that [INDUSTRY] must follow in [COUNTRY]. Include licensing authorities, export compliance requirements, required certifications for domestic sales, recent changes to regulations in the last two years, and any pending laws that could affect the sector.

With this prompt, we quickly got structured information on GMP certifications, AYUSH Ministry licensing, and FSSAI requirements for products meant for consumption.

• Competitive Landscape Analysis: Before judging a company’s position, you need to know who else is in the market. AI is great at pulling together public data into a clear picture.

Prompt template:

Make a map of the competitive landscape for [INDUSTRY] in [COUNTRY]. Find the top 10 players by market share or revenue, their positioning (premium, mass market, or export-focused), recent mergers and acquisitions, key differences between leaders, and any major foreign investments in the last three years.

This gave us a quick sense of where our target stood compared to others and even highlighted two recent foreign acquisitions we hadn’t known about.

• Scanning for Multiple Import Regulations: Our investment depended on GCC export potential, so I needed to learn about import rules for herbal medicines in the UAE, Saudi Arabia, Qatar, and Kuwait—all at once.

  
  

Prompt template:

Compare the rules for importing [PRODUCT CATEGORY] in [LIST OF COUNTRIES]. For each, list the regulatory body, registration or licensing needs, labelling requirements, restricted ingredients, typical approval timelines and any trade agreements that affect duties.

Instead of researching each market separately, AI gave us a single comparison—showing, for example, that Saudi Arabia’s registration rules were the strictest, which affected our timeline.

What AI Cannot Do: 2 Key Limitations

1. Understand Political and Cultural Realities:

   AI can tell you what the rules are, but not how things actually work in practice. In cross-border deals, success often comes down to the difference between what’s written and what really happens. Personal relationships, informal approvals, and how regulators feel about foreign ownership—all of this is missing from AI results. For our Ayurvedic project, really understanding the Indian government’s “Make in India” attitude towards foreign buyers meant talking directly with local advisors who know the landscape. No AI prompt can replace those real conversations.
   

2. Spot Customer Preferences and Local Market Details: AI is great for hard data like market size and basic regulations. But it can’t tell you what packaging GCC customers prefer for Ayurvedic products, whether Indian brands are seen as premium or budget in Gulf countries, or how sizing expectations differ between India and the GCC. To fill these gaps, we turned to traditional market research—focus groups and retail audits in our target markets. AI helped us ask better questions, but only real-world research gave us the answers we needed.

The Practical Framework

Here's how I now structure AI-assisted due diligence:

Phase	AI Role	Human Role
Initial screening	Regulatory mapping, competitive landscape, market sizing	Decision to proceed
Deep dive	Document review, cross-jurisdiction comparison	Relationship assessment, cultural context
Final evaluation	Data synthesis, gap identification	Judgment call, negotiation strategy

AI accelerates the "what" of due diligence. The "why" and "whether" still require human judgement, relationships, and boots-on-ground research.

Your Turn

If you’re looking at cross-border opportunities, try using the prompt templates above and tailor them to your own sector and target markets. I recommend starting with regulatory landscape mapping—it’s usually where AI can save you the most time right away.

Have you found any AI tools especially helpful when evaluating deals? Or have you run into roadblocks where AI doesn’t quite deliver? I’d love to hear what’s working for you and where things have been challenging. Let’s share ideas!