Insights

Human Oversight, Ethical Reasoning & AI Regulation

How can organisations demonstrate that human oversight works when AI influences their decisions?

At Blooming & Bloom, this question connects our consultancy approach with the development of DcGn®, our moral and ethical reasoning framework. We are exploring how structured ethical review could support the practical implementation of AI regulation and inform future policy discussions.

The Potential Contribution of DcGn®

Regulatory principles need practical methods through which organisations can examine decisions, assign responsibility and demonstrate their actions.

DcGn® is being developed to support this work in three areas.

Turning oversight principles into practical questions

Who reviews an AI recommendation? What information do they need? Can they challenge it, override it or stop its use?

DcGn® could help structure these questions within a review process, making responsibilities and gaps in oversight easier to identify.

Making ethical considerations explicit

AI-assisted decisions can involve competing concerns, including fairness, efficiency, privacy and individual autonomy.

DcGn® is designed to help people examine these tensions, consider affected stakeholders and explore alternatives. Its potential contribution is to make those considerations visible and open to scrutiny.

Supporting documented human judgment

We are exploring how DcGn® could help record the evidence considered, assumptions questioned, uncertainties identified and reasons behind a decision.

Such records could support organisational accountability and subsequent assessment. They would explain the human decision process, rather than reveal the internal reasoning of an AI model.

Building Evidence for Policy Contributions

DcGn® currently exists as a conceptual framework supported by a Coda-based working instrument. Its effectiveness and relevance to regulatory practice require further testing and independent evaluation.

Through defined pilots, comparative assessment and research collaboration, we aim to establish where the framework adds practical value.

Evidence from this work could inform:

Responses to relevant AI policy consultations.
Recommendations for implementing meaningful human oversight.
Case studies on ethical review and decision accountability.
Collaborative research into AI assurance methods.

These are intended areas of contribution, rather than claims of established regulatory influence.

An Independent, Transparent Position

DcGn® is not a regulatory standard, certification scheme or guarantee of compliance. Its development aims to complement established requirements and specialist expertise.

Blooming & Bloom does not represent a regulator or government body. We will distinguish evidence, interpretation and proposals in any research or policy contributions we publish.

Collaborate With Us

We welcome enquiries from researchers, organisations and policy communities interested in evaluating how structured ethical reasoning could contribute to responsible AI use and the implementation of AI regulation.