AI transparency and accountability: what governance teams are missing

TL;DR: Fairness, accountability and transparency problems in applied AI cannot be solved by explainable AI alone, according to Asenion's analysis of Stefan Larsson's socio-legal research, because biased training data, contested norms and human value mirroring create governance gaps that computer science methods do not close. The implication is that AI governance must combine technical controls with legal and normative scrutiny.

Editorial analysis by NHI Mgmt Group, based on content published by Asenion: “The socio-legal relevance of artificial intelligence”.

Key questions

Q: How should organisations govern AI systems that are explainable but still unfair?

A: Treat explainability as a supporting control, not proof of fairness.

Q: Why does biased training data matter if model performance looks good?

A: Good performance on benchmark tasks does not eliminate unfairness in real decisions.

Q: What are the best ways to make AI transparency operational?

A: Use a layered approach.

Practitioner guidance

• Define accountability for AI decisions Assign a named business owner, a technical owner and a review path for each AI use case so explanation, approval and challenge are not left implicit.
• Test training data for embedded bias Review dataset provenance, class balance and label quality before deployment, then revalidate when the model is retrained or repurposed.
• Map transparency controls across the lifecycle Document purpose, data sources, model assumptions, output use and appeal process so transparency is treated as an end-to-end control set.

Bottom line: Applied AI can be explainable and still fail governance tests if accountability, challenge rights and data provenance are weak.

Explore further

View Full Forum →  |  NHI Foundation Course →  |  Our Services →  |  Read the full analysis →