AI Transparency Statement
ALMA uses General-Purpose AI (GPAI) models to generate analytical content within the platform. In accordance with Article 50 of the EU AI Act, this page discloses which features use AI generation, how outputs should be interpreted, and how to provide feedback.
All AI-generated content within ALMA is clearly labelled with an AI-Generated Content badge at the point of first exposure. This badge links to this transparency statement. No AI-generated content is presented as human-authored.
Article 50 of the EU AI Act (Regulation (EU) 2024/1689) imposes transparency obligations on providers and deployers of AI systems that interact with natural persons or generate synthetic content. ALMA is subject to Article 50(2) (AI-generated text) and Article 50(5) (platform-level disclosure).
ALMA does not use AI systems for decisions that produce legal or similarly significant effects on individuals. All AI-generated content is advisory in nature and requires human review before use in formal compliance documentation, governance submissions, or regulatory filings.
The AI models used within ALMA are classified as General-Purpose AI (GPAI) models under Chapter V of the EU AI Act. ALMA acts as a deployer of these models and has implemented the transparency measures described on this page.
AI-Powered Features
The following features use AI generation. Each is manually triggered by an authorised user and requires human review before use in formal documentation.
Narrative Insights
Art. 50(2) — disclosure of AI-generated text.Generates a written narrative interpretation of your organisation's ALMA assessment results, including dimension-level commentary and thematic synthesis of free-text responses.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. Outputs are advisory and must be reviewed by a qualified professional before use in formal documentation.
Personalised Recommendations
Art. 50(2) — disclosure of AI-generated text.Produces a prioritised list of intervention recommendations calibrated to your organisation's score profile, industry sector, and size band.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. Recommendations are indicative and should be validated against your organisation's specific context.
Reflective Response Analysis
Art. 50(2) — disclosure of AI-generated text.Analyses anonymised free-text responses from Section G of the ALMA assessment to identify recurring themes, sentiment patterns, and organisational concerns.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. Thematic analysis is probabilistic and may not capture all nuances of qualitative data.
Board-Ready Report
Art. 50(2) — disclosure of AI-generated text; Art. 50(4) — human editorial control gate.Generates an executive summary document containing overall ALMA scores, top findings, EU AI Act article readiness indicators, and upcoming milestones.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. A human review acknowledgement is requested before download for use in formal governance submissions.
Compliance Evidence Pack
Art. 50(2) — disclosure of AI-generated text; Art. 50(4) — human editorial control gate.Compiles a structured compliance documentation package for regulatory submission, including dimension scores, EU AI Act article compliance mapping, action plans, and a formal declaration.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. A human review acknowledgement is requested before download. The declarant confirms the content has been reviewed by a qualified professional.
ISO 42001 Gap Analysis
Art. 50(2) — disclosure of AI-generated text.Produces a gap analysis report mapping ALMA assessment results to ISO/IEC 42001:2023 clauses and EU AI Act obligations, identifying conformity gaps and recommended actions.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. Gap analysis outputs are indicative and do not constitute formal certification advice.
AI Insights Engine
Art. 50(2) — disclosure of AI-generated text.Provides dimension-level narrative interpretation and trend analysis across multiple assessment cycles, surfacing longitudinal patterns in AI readiness and governance maturity.
Trigger
Manually triggered by a Manager or Admin user.
Model Type
General-Purpose AI (GPAI) — large language model.
Human Review
Required. Trend interpretations are based on aggregated assessment data and should be contextualised by qualified advisors.
- 1All AI-generated outputs are labelled at the point of first exposure with an 'AI-Generated Content' badge.
- 2Board Reports and Compliance Evidence Packs require an explicit human review acknowledgement before download for use in formal submissions.
- 3No AI-generated content is used for automated decision-making that produces legal or similarly significant effects on individuals.
- 4AI generation is always manually triggered by an authorised Manager or Admin user — no autonomous generation occurs.
- 5All AI generation events are logged server-side with feature name, timestamp, and organisation identifier (no personal data) for a minimum of 12 months.
- 6ALMA does not use AI to assess, score, or classify individual respondents. Scoring is deterministic and based on validated psychometric algorithms.
If you have concerns about AI-generated content produced by ALMA, or wish to report an inaccurate, misleading, or harmful output, please contact us at [email protected]. We aim to respond within 5 business days.
ALMA is developed and operated by The Responsible AI Center. Mulya van Roon, Founder & Principal Advisor, is the designated AI transparency contact.
This transparency statement was last updated on 26 March 2026.