Product Demo Narrative & Demo Script

Demo Objective The purpose of the demo is to show how Aegis SafeForge helps safety and cybersecurity engineering teams use AI in a controlled, traceable, and reviewable workflow. The demo should make one message clear: Aegis SafeForge is not a generic AI chatbot. It is a governed engineering workspace where AI supports structured artifact generation, while engineers remain in control of review, approval, traceability, and evidence. Recommended Demo Flow Set the context: explain that functional safety and cybersecurity teams still spend too much time maintaining connected artifacts manually across Excel, Word, PDFs, and disconnected tools. Introduce the platform: Aegis SafeForge is an AI-assisted engineering workspace for functional safety and cybersecurity workflows. Show the 2-3 minute workflow video: let the video play through once, then explain the workflow afterward. Explain the four layers: knowledge layer, artifact layer, governance layer, and traceability layer. Transition to pilot discussion: propose validating one focused workflow alongside the partner’s current process. Video Voiceover / Caption Structure

Live Demo Script After the Video What you just saw is the current Aegis SafeForge workflow. The important point is that AI is not operating as an uncontrolled document generator. It is embedded inside a structured engineering workflow. There are four core layers in the platform: Knowledge layer: Teams upload project-specific documents, and the AI can use those sources during generation. Artifact layer: Work products such as HARA, TARA, Safety Goals, and requirements are stored as structured records, not only as free-text documents. Governance layer: AI-generated outputs enter the system as suggestions and must be reviewed, edited, approved, or rejected by engineers. Traceability layer: The platform links upstream and downstream artifacts so teams can understand how engineering decisions connect across the workflow. The value for a pilot partner is that we can test this on a real or representative workflow and measure whether it reduces first-pass documentation effort, improves consistency, and makes review and traceability easier. Short Version for a 3-Minute Demo Call Aegis SafeForge is an AI-assisted engineering workspace for functional safety and cybersecurity teams. The problem we are solving is that teams still spend too much time creating and maintaining connected artifacts manually across Excel, Word, PDFs, shared folders, and legacy tools. With Aegis SafeForge, teams can upload project-specific knowledge sources, generate structured artifacts such as HARA, TARA, Safety Goals, and requirements, review AI suggestions, approve or reject outputs, and preserve traceability across the workflow. The important distinction is governance: AI-generated outputs are not treated as final decisions. They move through lifecycle states such as AI Suggested, Edited, Needs Review, and Approved. The platform is currently ready for pilot and design partner validation. We are looking to test real workflows, validate export needs, refine traceability requirements, and understand which deployment and security requirements matter most for enterprise adoption. Questions to Ask Pilot Partners After the Demo Which parts of your current safety or cybersecurity workflow are still managed manually? Where do you lose the most time today: artifact creation, review, traceability, updates, or evidence preparation? Do your teams currently use Excel, Word, legacy tools, or a mix of systems? Would AI-assisted first-pass generation be useful if engineers remain fully in control of review and approval? Which workflow would be most valuable to test first: HARA, TARA, Safety Goals, requirements, or evidence packaging? What export formats would you need for the platform to fit your current process? What security, deployment, or data isolation requirements would matter for a pilot? What would a successful pilot outcome look like for your team? Who else in your organization would need to evaluate a tool like this? Would you be open to defining a small design partner pilot around one representative workflow? Suggested Pilot Transition Based on what we have shown, we would not suggest starting with a large rollout. The best next step would be a focused design partner pilot around one workflow, for example Item Definition to HARA to Safety Goals, or a TARA workflow. The goal would be to compare the current manual process against Aegis SafeForge and evaluate time saved, artifact quality, review effort, traceability, and export usability. Suggested Pilot Scope Slide Text One representative system, item, or product workflow. Upload of selected project or sample documents. AI-assisted generation of initial artifacts. Engineer review and editing. Traceability validation. Export validation. Feedback session with safety or cybersecurity engineers. Summary of findings and next-step recommendations. Key Messages to Repeat AI proposes, engineers review. The platform is designed for governance, not uncontrolled generation. Traceability is built into the workflow. Artifacts are structured records, not just text documents. The goal is to reduce repetitive work without losing engineering control. We are looking for design partners to validate this in real workflows. The pilot can run alongside existing processes instead of replacing them immediately. What Not to Overclaim

Demo Closing Statement Aegis SafeForge is being built for engineering teams that want to use AI, but cannot rely on uncontrolled AI outputs for safety or cybersecurity work. Our focus is governed AI: source-grounded, structured, reviewable, traceable, and export-ready. We are now looking for pilot and design partners to validate the platform in real workflows and help shape the product around practical engineering needs.

Appendix - Suggested Online Knowledge Base Mapping The documents in this packet can be converted into an online knowledge base so pilot partners can be directed to controlled web pages instead of receiving PDFs. Recommended starting structure:

Recommended access model: keep public marketing content on the main website, keep semi-public pilot documentation behind a lightweight access gate or unlisted docs area, and keep legal/IP/financial documents in a controlled data-room tool with access logging.