Digital DNA (DDNA): A Practical Way to Keep AI Grounded in the Right Context
This short explainer introduces Digital DNA (DDNA) – a lightweight, reality-aligned way to carry who the user is, what really matters right now, and where the guardrails are across all the AI tools they use.
Instead of each system guessing context from scratch (and quietly drifting into pseudo-friend or pseudo-therapist roles), DDNA treats context like “digital genetics”: a structured, portable profile that can be read by different AI systems but remains under the user’s control. The paper shows:
- how DDNA encodes roles, boundaries and purpose in a compact schema,
- how it can reduce misalignment and relational drift when people juggle multiple AIs,
- concrete examples for education, mental health, and everyday productivity, and
- how DDNA can plug into existing AI governance and risk-management work without rewriting everything.
The explainer is meant as an invitation to policymakers, governance practitioners, and builders to explore DDNA as a practical complement to the EU AI Act: a way to make “user context and safeguards” real, portable, and testable across systems.
Zenodo DOI: 10.5281/zenodo.17842920
Author ORCID: 0009-0008-1764-4108
Contact: niels.bellens@proton.me
- Žymos
- discussion ai regulation
- Prisijunkite, kad galėtumėte skelbti komentarus.