Legacy Systems vs AI: A Recipe for Failure
40% of AI projects fail because of legacy systems. Here's how not to fall into that trap.
You've got a great AI idea. The team is motivated. The budget is approved. And then you discover that your core system has an API from 2008 and returns data in a format nobody remembers anymore.
Gartner predicts that more than 40% of agentic AI projects will fail by the end of 2027. Why? Costs keep climbing, and the new systems don't fit what the company already runs. Most people treat this as a technical problem. It's really a strategic one.
Why does legacy kill AI?
AI needs three things: data, speed and flexibility. Legacy systems offer the opposite of each.
TYPICAL LEGACY PROBLEMS
- Data in silos: ERP doesn't talk to CRM, CRM doesn't talk to WMS
- No real-time: Batch processing once a day, while AI needs the data right now
- Outdated APIs: SOAP, FTP, or, worse, no API at all
- No documentation: The only person who understood the system has retired
3 integration strategies
1. Abstraction layer (Data Mesh)
Instead of touching the legacy, build an intermediate layer. An API Gateway that translates old formats into new ones. A Data Lake that pulls data from different sources together. Zhamak Dehghani called this Data Mesh: every dataset has a real owner, and you treat it like a product.
Pros: You don't touch the legacy. Cons: One more layer to maintain, plus cost.
2. Strangler Fig Pattern
Martin Fowler described this pattern: piece by piece, you replace legacy functionality with new microservices. AI runs on the new services while the old ones slowly die off. The name comes from the strangler fig, which grows over its host and replaces it.
Pros: Lower risk. Cons: Slow. Years, not months.
3. RPA + AI Bridge
You use RPA to "scrape" data from the legacy UI, AI processes it, and RPA writes the results back. It sounds ugly, but it works.
Pros: Fast. Cons: Brittle, needs maintenance.
"The wrong question is 'how do I integrate AI with the whole legacy system'. The better one: 'what's the minimum data access I need to get AI working'. Usually you need far less than you assume."
This is what I do hands-on — advising on AI strategy and building agents that survive the demo.
Audit before you start
Before you start an AI project, answer these questions:
LEGACY READINESS CHECKLIST
- □ Is the data I need available through an API?
- □ What's the latency of data access?
- □ Can I write AI results back into the system?
- □ Who is the technical owner of the legacy system?
- □ What are the maintenance and downtime windows?
- □ Is there integration documentation?
Case study: A client with a 15-year-old ERP
A manufacturing company wanted to deploy AI for demand forecasting. The problem: their ERP had no API, only a CSV export once a day.
The solution:
- • A script that automatically pulls the CSV at 6 a.m.
- • ETL into a Data Lake (PostgreSQL)
- • An AI model trained on yesterday's data
- • Predictions emailed to the planners
It's not a textbook architecture. But it works and it delivers value.
Your first move
Pick one process that genuinely hurts. Check whether you can pull even a daily CSV export out of the legacy. If you can, that's enough to stand up a first model in two weeks and show real numbers. The big rewrite can wait.
