Aridio Silva

Independent Researcher | AI Systems Architecture | Agentic AI | Edge AI | AI Security & Governance

Background

My professional career in software development began in 1977, providing a long-term perspective on the evolution of software architecture, distributed systems, security engineering, and Artificial Intelligence.

Today, my research focuses on the architectural and governance challenges introduced by autonomous and distributed AI systems.

Independent Researcher | AI Systems Architecture | Agentic AI | Edge AI | AI Security & Governance

I am an independent researcher and technology professional based in Brazil, with a research focus on the architecture, security, governance, and trustworthiness of distributed and autonomous Artificial Intelligence systems.

My current work explores the convergence of Agentic AI, Multi-Agent Systems, Edge AI, Zero-Trust Architecture, AI Security, Governance, Risk and Compliance (GRC), and Security-by-Design.

I am particularly interested in how increasingly autonomous AI agents can operate across distributed and edge environments while remaining subject to explicit, bounded, auditable, and revocable authority.

Current Research

I am the creator and lead researcher of SGAEIA — Secure Governed Autonomous Edge Intelligence Architecture, an open and vendor-neutral reference architecture for governed distributed multi-agent systems.

SGAEIA investigates the transition from conventional Edge AI toward Governed Autonomous Edge Intelligence, where AI agents may perceive, reason, communicate, delegate, and act across heterogeneous edge-cloud infrastructures while remaining constrained by enforceable security and governance policies.

The architecture explores concepts including:

• bounded and revocable agent authority;
• authenticated and traceable delegation;
• Zero-Trust runtime enforcement;
• multi-agent trust boundaries;
• continuous Governance, Risk and Compliance;
• Security-by-Design, Security-First, and Shift-Left engineering;
• threat modeling using STRIDE and OWASP principles;
• formalized security invariants;
• Evidence-as-Code and continuous assurance;
• security-preserving technology substitution;
• distributed and hierarchical Edge AI architectures.

Research Interests

My broader research interests include Agentic AI Security, Multi-Agent Systems, Edge Intelligence, Distributed AI, Zero Trust, AI Governance, trustworthy autonomous systems, cyber-resilience, secure AI architectures, and continuous security assurance.

I am especially interested in the architectural question:

How can autonomous AI systems be granted enough authority to perform useful actions while ensuring that such authority remains bounded, verifiable, auditable, and revocable?

Open Research

I believe that security and governance should be treated as architectural properties of autonomous AI systems rather than as controls added after deployment.

My work therefore follows an open research and reproducible engineering approach, combining conceptual architecture, formal specifications, threat models, security invariants, executable evidence, reference implementations, and persistent research artifacts.

SGAEIA

Secure Governed Autonomous Edge Intelligence Architecture

GitHub: https://github.com/aridiosilva/SGAEIA

Archived research artifact: https://doi.org/10.5281/zenodo.22557796

Research Identity

Aridio Silva Independent Researcher — Brazil

Research Profiles

Aridio Silva
Independent Researcher — Brazil

• ORCID: https://orcid.org/0009-0008-2411-6995
• Google Scholar: https://scholar.google.com/citations?user=rPn5O48AAAAJ
• GitHub: https://github.com/aridiosilva
• Zenodo: https://doi.org/10.5281/zenodo.22557796
• LinkedIn: https://www.linkedin.com/in/aridio-silva-74997111/
• SGAEIA: https://github.com/aridiosilva/SGAEIA
• medium.com https://medium.com/@aridiosilva