San Francisco • October 2026
By the Cerebital Team
AI at the Core of the Product Lifecycle
The Cerebital team attended The AI Conference 2026 in San Francisco, a global gathering where researchers, developers, and industry leaders explored how artificial intelligence is evolving from experimentation to enterprise infrastructure.
This year’s theme was clear: operationalizing AI responsibly, efficiently, and at scale.
For Cerebital and our PLM platform, Nora IPLM, this mission resonates deeply. The discussions reinforced our vision of an AI-driven PLM ecosystem that connects design, data, and decisions into a continuous cycle of intelligence.
What Does AI in PLM Mean?
AI in Product Lifecycle Management applies artificial intelligence to connected product data and lifecycle processes such as BOMs, parts, requirements, documents, engineering changes, suppliers, projects, and workflows.
Instead of using PLM only to store and govern information, AI can help teams retrieve product knowledge, summarize complex structures, identify patterns and risks, explain change impact, detect reuse opportunities, and prepare structured outputs for human review.
Agentic AI: The Rise of Digital Engineering Teammates
A defining trend this year was the emergence of Agentic AI, systems capable of reasoning, planning, and acting autonomously.
These intelligent agents are moving beyond simple prompt-response patterns to orchestrate workflows, retrieve information, and support decision-making.
In PLM environments, they are evolving into digital teammates that can:
• Suggest design improvements based on historical performance and material data
• Monitor dependencies across product structures and flag risks early
• Coordinate engineering change processes automatically
At Nora IPLM, we see Agentic AI not as a replacement for engineers but as a collaborative force that accelerates innovation and enhances design intelligence.
Within PLM environments, such systems are transforming into digital teammates that can:
• Suggest design improvements based on past performance and materials data
• Monitor dependencies across product structures and flag potential risks
• Coordinate engineering change processes automatically
At Nora IPLM, Agentic AI is being developed not as a replacement for engineers, but as a collaborative force multiplier that enables faster, smarter, and more informed design cycles.
From AI Assistant to Lifecycle-Aware AI Agent
A general AI assistant responds primarily to the information included in a prompt. A lifecycle-aware PLM agent can work with structured product context: the active BOM, selected item, revision, change record, workflow state, documents, relationships, and user permissions.
That distinction matters in engineering because useful AI needs to understand not only text, but also:
- Which product object is being reviewed
- Which revision and lifecycle state is current
- How parts, BOMs, documents, requirements, and changes are related
- Which downstream objects may be affected by a decision
- What information the current user is permitted to access
- Where human review or formal approval is still required
Why Product Context Matters More Than Generic AI Knowledge
Engineering questions rarely exist in isolation. A part can belong to several assemblies, have multiple revisions, depend on supplier information, appear in open changes, and be governed by different lifecycle states.
Retrieval-Augmented Generation can ground an AI response in enterprise information, while graph-based product knowledge preserves the relationships between objects. Together, these approaches can help an AI system answer not only “What does this document say?” but also “What is connected to this object, what changed, and what could be affected?”
This is especially relevant for impact analysis, where the value comes from preserving the evidence behind a recommendation rather than producing an isolated natural-language answer.
Knowledge Graphs + RAG: Building the PLM Digital Brain
Retrieval-Augmented Generation (RAG) and Knowledge Graphs were standout topics, presented as the foundation of enterprise AI.
RAG allows large language models to access live organizational knowledge, while Knowledge Graphs structure relationships between data, designs, and decisions.
Together, they form a digital brain that connects every element of product development, enabling AI to answer questions such as:
“Which past designs used the same alloy and failed under vibration testing?”
Cerebital is advancing this vision by integrating RAG and graph-based reasoning directly into Nora IPLM and creating a platform that continuously learns, adapts, and improves.
RAG allows large language models to access live organizational knowledge, while Knowledge Graphs capture structured relationships between data, designs, and decisions.
In the context of PLM, this combination forms a digital brain that connects every element of product development.
It enables AI to answer queries like:
“Which past designs used the same alloy and failed under vibration testing?”
Cerebital is advancing this vision by integrating RAG and graph-based reasoning directly into Nora IPLM, creating a platform that learns and adapts with every iteration.
Multimodal Intelligence: When AI Understands 3D
Beyond text and code, multimodal AI took center stage, showcasing how models now interpret 3D geometry, images, and documents together.
This breakthrough is critical for engineering and manufacturing, where design intent often resides within CAD models and technical drawings.
For PLM users, multimodal intelligence enables:
• Visual identification of duplicate or similar parts across assemblies
• Manufacturability assessment directly from geometric features
• Automatic generation of process plans and instructions from 3D models
These capabilities align with Cerebital’s work in spatial AI and shape-based reasoning, helping PLM systems truly “understand” the products they manage.
Multimodal AI for Engineering Data
Multimodal AI is particularly relevant to PLM because engineering knowledge is distributed across different data types: structured attributes, BOM hierarchies, CAD geometry, drawings, specifications, images, test results, and documents.
The long-term opportunity is to analyze these sources together while preserving product identity and lifecycle context. For example, a similarity workflow could combine geometric resemblance with material, dimensions, supplier, cost, revision history, and where-used information before an engineer decides whether a part can be reused.
Responsible AI: Governance as a Core Principle
Ethics, transparency, and traceability were consistent themes throughout the conference.
As AI becomes an enterprise decision partner, governance and explainability must be integral, not optional.
For PLM systems, responsible AI means every insight should be:
• Logged and auditable
• Linked to its data sources
• Validated by human oversight when necessary
Nora IPLM is built with governance by design, embedding compliance and accountability into every layer of its intelligence framework.
In engineering, trust is the true foundation.
Infrastructure & Efficiency: Scaling Intelligence Sustainably
Another focus area was AI infrastructure. The conversation is shifting from ever-larger models toward smaller, specialized systems that are faster, more secure, and cost-effective.
For enterprise PLM, this translates to:
• Faster inference through optimized models
• Hybrid or on-premises deployments to safeguard IP
• Scalable architectures balancing performance and privacy
Cerebital is building toward this future by delivering lightweight, high-performance AI that scales with the customer’s data, not against it.
AI Use Cases Across the Product Lifecycle
PLM Area | AI-Assisted Task | Potential Value |
BOM Management | Summarize BOM health, component risks, cost and readiness | Faster product-structure reviews |
Explain affected objects, dependencies and open decisions | Better-informed change reviews | |
Parts & Reuse | Find similar or duplicate components | More reuse and less part proliferation |
Documents | Extract fields, summarize content and flag missing information | Less manual document review |
Requirements | Review coverage and identify gaps or relationships | Improved requirement traceability |
Projects & Workflows | Summarize status, bottlenecks and readiness signals | Faster operational review |
Product Knowledge | Answer contextual questions across connected objects | Faster engineering investigation |
How This Conference Vision Is Appearing in Nora IPLM
The conference themes described in this article now map to concrete areas of Nora IPLM’s AI direction. Nora Prima is positioned as a lifecycle-aware AI layer that works with BOMs, changes, documents, suppliers, workflows, revisions, and other connected product context.
Current Nora IPLM AI capabilities and product direction include BOM analysis and health summaries, part similarity and reuse, document intelligence, contextual product-data queries, lifecycle summaries, change explanations, risk and readiness analysis, and repeatable prompt templates for common PLM work.
Nora IPLM’s Knowledge Thread also extends the knowledge-graph idea by representing PLM objects and their typed relationships as an explorable product-knowledge network. This supports relationship investigation, evidence-aware paths, impact simulation, alternate-item comparison, and contextual Prima AI analysis while keeping the underlying product evidence visible.
A New Era for Product Lifecycle Intelligence
The conversations in the conference reaffirmed what we have been building toward: AI is no longer experimental; it is essential.
For PLM, this shift transforms the platform from a static repository into a living, learning system that predicts, collaborates, and evolves with every decision.
Through Nora IPLM, Cerebital is leading this transformation, where AI agents think with data, knowledge graphs connect enterprise context, and multimodal intelligence brings design understanding to life.
The Road Ahead
AI is not replacing creativity; it is amplifying it.
The next generation of PLM will be measured not only by how it manages information but by how intelligently it learns from it.
At Cerebital, we are building that future today by turning product lifecycle management into product intelligence management.
See the consequence before you commit to the change
Discover how Nora IPLM Knowledge Thread turns connected product data into a traceable AI Change Impact Simulation, from qualified evidence and visual propagation to alternate-item, carbon, and AI-assisted analysis.
