Introduction
How Will AI Truly Change PLM in 3 Years? – It’s a question that’s been on the minds of many PLM professionals. In recent months, I’ve come across a wide range of thoughtful perspectives from industry leaders and solution providers on how Artificial Intelligence (AI) is poised to reshape Product Lifecycle Management. Many of these insights are forward-looking and visionary, and they’ve inspired me to reflect on a simple, practical question:
What will AI truly change in the daily life of a PLM user over the next three years?
Drawing from real-world implementations and challenges faced by cross-functional NPD/NPI teams, this blog explores how AI, especially large language models and agentic AI, can enhance what PLM already does well by removing friction, improving usability, and enabling smarter decisions.
This isn’t about hype. It’s about value where it matters most.
What Problems Still Exist? (Even with current PLMs in the market)
Let’s be clear: modern PLM systems already solve core challenges. However, execution gaps still exist, especially in high-velocity product development.
Here are 3 recurring day-to-day issues I see even in mature PLM environments:
Circular References or Misconfigurations
Users (especially new or cross-functional team members) often struggle with dependencies or relationships in structures like BOMs or CAD models. Errors like circular references are caught late or resolved manually with help from experts.
Searching for the “Right” Version
Whether it’s a BOM, spec, drawing, or change order, finding the right version for your role and task is still harder than it should be. This slows down decisions and invites rework.
System Fatigue & Collaboration Breakdown
Engineers, sourcing leads, and quality teams often feel the system works against them, leading to over-reliance on email or offline tracking. The PLM data is there, but the user experience gets in the way.
These aren’t system flaws, they’re experience gaps. And this is where AI can help.
What Can AI Realistically Do for PLM by 2028?
Let’s skip the buzzwords and focus on the value-added enhancements that AI can bring, not by reinventing PLM but by making it easier and smarter to use.
AI Will Enhance What Already Works
PLM is not broken. It handles product data, change control, workflows, and traceability well. What AI will do is simplify access, personalize responses, and guide users through complex actions, reducing dependency on experts and documentation.
LLMs as Your PLM Copilot
LLMs can convert plain English into precise queries inside the PLM system.
For example:
“Show me the release-ready BOM for Project abc.”
“Which change requests are pending for approval from Quality?”
No clicks. No search filters. Just answers in context, based on role and permissions.
Agentic AI: Guided Problem-Solving
Consider a frequent issue like a circular reference in a BOM or CAD assembly.
Rather than just flagging it, AI will:
Explain the problem in plain language.
Suggest how to fix it using system rules and past examples.
Walk the user through the solution or fix it if permissions allow.
This isn’t automation, it’s self-guided PLM resolution, increasing confidence and speed.
Role-Aware Search & Decision Support
Let’s say a Production Planning & Control user searches for a BOM.
An AI-powered PLM system will:
Detect the role.
Display only the relevant BOM (e.g., post-NPI approved version for ERP release).
Highlight pending inputs or dependencies the PPC role cares about.
This is role-based intelligence, not generic search. It’s how AI brings relevance to PLM.
The Value Shift: From Information Storage to Intelligent Enablement
PLM has evolved from being a system of record to a platform for orchestrating product development processes. But AI takes it a step further, from structured systems to contextual support environments.
Yesterday’s PLM: Centralized and rule-driven, but rigid.
Today’s PLM: Smart Connected PLM.
Future-ready PLM: Augmented by AI will act as intelligent copilots, not only detecting such issues early but also learning from each resolution to guide users across roles toward faster, smarter fixes.
Importantly, AI doesn’t replace PLM, it elevates it. It removes friction, improves onboarding, reduces training needs, and empowers all users to participate meaningfully.
Conclusion: The Human-Product Core Still Matters Most
AI, LLMs, and intelligent agents will reshape how we use PLM, but they won’t replace its foundations. In the end: Great PLM is still about People, Process, and Product. Without alignment across teams, clear responsibilities, and a shared vision of product success, even the smartest AI will only amplify confusion.
So, as we embrace AI in PLM, let’s stay focused on what matters:
- Empowering humans.
- Respecting product integrity.
- Simplifying processes.
Because when AI works in the service of the product and the people who build it, that’s when digital transformation truly adds value.
Inspired by conversations with global peers, NPD professionals, and system thinkers who live and breathe PLM every day.
