One-Page Company Summary
What We Are Building Aegis SafeForge is an AI-assisted engineering platform for regulated safety and cybersecurity workflows. The platform helps engineering teams create, review, trace, and manage structured artifacts such as HARA, TARA, Safety Goals, requirements, and compliance evidence. It combines AI-assisted generation with human review, structured governance, source grounding, traceability, and export-ready documentation. Aegis SafeForge is not a generic chatbot or document generator. It is designed as an engineering workspace where AI acts as a controlled co-pilot and engineers remain responsible for review, approval, and final decisions. Problem Functional safety and cybersecurity workflows are still heavily manual. Teams rewrite the same information across multiple artifacts. Traceability between hazards, risks, goals, requirements, and evidence is difficult to maintain. Reviews are slowed by long documents, fragmented sources, and disconnected tools. Generic AI tools can help draft content but usually lack governance, source grounding, review states, and auditability. Solution Aegis SafeForge provides a structured, AI-assisted workspace for safety and cybersecurity engineering. The platform supports project-scoped knowledge ingestion, AI-assisted artifact generation, review workflows, traceability, real-time collaboration, role-based access, and compliance-oriented export. Core principle: AI proposes. Engineers review. The platform preserves traceability, evidence, and control. Current Platform Capabilities AI-assisted generation of HARA, TARA, Safety Goals, and structured safety/security artifacts. Project-scoped knowledge source ingestion for PDFs, DOCX files, spreadsheets, and evidence files. Source citations, reasoning, and confidence indicators for AI-generated outputs where available. Lifecycle states such as AI Suggested, User Created, Edited, Approved, and Needs Review. Traceability engine linking upstream and downstream artifacts. Requirements extraction and HLR/LLR management. Industry profile support for automotive, railway, medical, and process safety domains. Real-time collaboration with presence tracking and field-level locking. Role-based access control across organizations, teams, and projects. Usage metering and credit tracking for AI usage. Excel exports and evidence package generation. SaaS web application and offline-capable Electron desktop delivery path. Initial Use Cases Item Definition to HARA generation. HARA to Safety Goals. Safety Goals to requirements and evidence structures. TARA and cybersecurity risk chains. Requirements extraction and baselining. Safety/security artifact review and approval. Traceability management across work products. Compliance evidence package preparation. What We Are Looking For Aegis SafeForge is looking for pilot and design partners who want to help validate the platform in real engineering workflows. Ideal partners include functional safety consultancies, cybersecurity engineering consultancies, Tier 1 suppliers, OEM innovation or engineering teams, safety managers, systems engineering teams, and organizations currently using Excel-heavy or document-heavy safety workflows.
title: Problem Statement & Solution Overview status: Draft audience: Public lastUpdated: 2026-06-08 visibility: Public
Problem Statement Functional safety and cybersecurity engineering are becoming more complex, but much of the daily workflow is still handled through manual, fragmented, and document-heavy processes. Engineering teams working on standards such as ISO 26262, ISO/SAE 21434, IEC 62278, ISO 14971, and IEC 61511 need to create and maintain many connected work products. These may include Item Definitions, HARA, TARA, Safety Goals, Cybersecurity Goals, requirements, verification evidence, review records, and compliance packages. In practice, much of this work is still managed through a combination of Excel files, Word documents, PDFs, email threads, shared folders, legacy tools, and manual review meetings. Engineering teams are expected to move faster, but the tools and processes used to manage safety and cybersecurity artifacts are often slow, repetitive, and difficult to keep consistent. Why This Problem Matters Safety and cybersecurity artifacts are not isolated documents. A change in an Item Definition can affect HARA. A change in HARA can affect Safety Goals. A change in Safety Goals can affect requirements. A change in requirements can affect verification evidence. A change in system architecture can affect both safety and cybersecurity analysis. Missing or outdated traceability links. Repeated rewriting of the same information across artifacts. Slow review and approval cycles. Inconsistent terminology across documents. Difficulty identifying which downstream work products are affected by upstream changes. Poor visibility into why a decision was made. Heavy effort to prepare audit-ready evidence. Difficulty scaling safety and cybersecurity work across multiple projects. The Manual Workflow Gap Many teams still rely on flexible but disconnected tools. Excel is familiar and easy to customize, but it becomes difficult to manage when projects grow in complexity. Word and PDF documents are useful for reporting, but they are not ideal as the primary system of record for structured engineering artifacts. Legacy tools can provide structure, but they are often expensive, complex, difficult to adapt, or not designed around AI-native workflows. Teams need the flexibility of Excel, the structure of engineering lifecycle tools, the acceleration of AI, and the governance required for regulated domains. The AI Adoption Problem Generative AI can help safety and cybersecurity teams draft, summarize, classify, and structure engineering information. However, generic AI tools alone are not sufficient for regulated engineering workflows. Outputs may not be grounded in project-specific source documents. Reasoning may not be traceable to evidence. Generated content may not follow approved schemas. There may be no review or approval lifecycle. There may be no link between generated artifacts and upstream engineering decisions. There may be no audit trail showing who reviewed or changed the output. Sensitive engineering data may be exposed to unsuitable environments. Solution Overview Aegis SafeForge is an AI-assisted engineering workspace for functional safety and cybersecurity teams. The platform helps teams generate, review, trace, and manage structured artifacts across safety and cybersecurity workflows. It combines AI-assisted generation with project-scoped knowledge retrieval, human review, lifecycle states, role-based access, traceability, real-time collaboration, and export-ready documentation. Aegis SafeForge does not replace safety or cybersecurity engineers. Instead, it reduces repetitive manual work and helps engineers manage complex artifact relationships more consistently. Governed AI Instead of Uncontrolled AI In Aegis SafeForge, AI-generated content is not treated as final truth. It enters the system as a proposal. Artifacts can move through lifecycle states such as AI Suggested, User Created, Edited, Needs Review, and Approved. This gives teams a clear distinction between machine-generated suggestions and engineer-approved project records. Traceability by Design Aegis SafeForge stores engineering work products as structured records rather than only as free-text documents. This allows the platform to maintain explicit links such as Item Definition to HARA, hazardous event to Safety Goal, Safety Goal to requirement, requirement to verification evidence, asset to threat scenario, and cybersecurity goal to requirement or control. Project-Scoped Knowledge Grounding Aegis SafeForge uses project-scoped knowledge retrieval to ground AI assistance in the user’s own engineering context. Each project can have its own source documents and vector collection. Users can control which documents are available to the AI retrieval context, improving relevance and visibility into the source basis for generated suggestions. Designed for Enterprise Engineering Teams Modular frontend, backend, and AI service architecture. Organization, team, and project hierarchy. Role-based access control. Project-scoped data isolation. Real-time collaboration and field-level locking. Review queues and lifecycle states. Usage metering and structured audit logs. SaaS and offline-capable desktop delivery path. Future support for on-prem or customer-controlled deployment models. Summary Aegis SafeForge addresses a clear gap in regulated engineering workflows. Teams need to move faster, but they cannot compromise traceability, evidence quality, review control, or engineering accountability. Generic AI tools can generate content, but they do not provide the structured governance needed for functional safety and cybersecurity work. Aegis SafeForge combines AI-assisted generation with structured engineering governance.
Current Platform Capability Update - July 2026
The platform now has a stronger compliance-readiness story than the original packet draft. In addition to safety and cybersecurity artifact workflows, ASF now includes pilot-level CRA compliance-engine records, lifecycle cases, release-package readiness, SBOM/dependency evidence support, vulnerability-management workflows, and CRA gap-report export.
This does not change the core message that AI proposes and engineers review. It sharpens it: ASF is being built to connect engineering work products to reusable evidence and compliance-readiness records, while keeping formal decisions, approvals, and legal conclusions under human control.