AI POSITIVE
Compliance is the floor. AI Positive is the ceiling.
The EU AI Act tells you what you must do. AI Positive tells you what you should do, and measures it.
Most AI regulation frameworks stop at compliance: are you legal? AI Positive goes further. It measures how much positive value your AI system creates: for users, for your organization, and for society. It is a performance standard, not a checklist.
Definition
AI Positive is an ethical performance scoring framework for AI systems, developed by Sprinkling Act as a complement to EU AI Act compliance (Regulation 2024/1689). It evaluates an AI system across 5 criteria (Governance & Accountability, Bias Reduction & Fairness, Explainability & Transparency, Net Human Impact, Control & Reversibility), each scored 0–20 for a 100-point total. Score thresholds map to 4 tiers: AI Positive Gold (80–100), AI Positive Verified (60–79), AI Positive Emerging (40–59), Not eligible (0–39). AI Positive is sequenced AFTER regulatory classification — compliance comes first. The framework synthesises the highest standards from EU AI Act, NIST AI RMF, OECD AI Principles, and UNESCO AI Ethics Recommendation.
WHAT IS AI POSITIVE
A framework for ethical AI performance.
AI Positive is a structured scoring framework that evaluates AI systems across 5 criteria. Each criterion is scored from 0 to 20, for a total of 100 points. The score measures not whether you comply with the law, but how well your AI performs on dimensions that regulations worldwide encourage but do not always mandate.
The criteria are grounded in the full spectrum of official AI regulation available: EU (AI Act, GDPR, AI Office guidance) and non-EU (NIST AI RMF, OECD AI Principles, UNESCO AI Ethics Recommendation, national frameworks). AI Positive is not limited to one jurisdiction. It synthesizes the highest standards from all.
AI Positive is designed to be complementary to regulatory classification. You cannot obtain an AI Positive score without first establishing your regulatory position through a Full Report. Compliance comes first, always.
The 5 criteria are designed in alignment with the spirit of the UN Sustainable Development Goals: because responsible AI and sustainable development are inseparable.
THE 5 CRITERIA
What we measure.
Governance & Accountability
0–20 pointsDoes someone own the outcomes? Is there a clear chain of responsibility from development to deployment? Are decisions documented and auditable?
Bias Reduction & Fairness
0–20 pointsHas the system been tested for discriminatory outcomes? Are demographic biases identified, measured, and mitigated? Is fairness defined and tracked over time?
Explainability & Transparency
0–20 pointsCan a non-technical stakeholder understand why the system produced a given output? Is the logic documented? Are limitations clearly communicated?
Net Human Impact
0–20 pointsDoes the system create more value for humans than it displaces? Is the impact on employment, autonomy, and wellbeing assessed and positive?
Control & Reversibility
0–20 pointsCan a human override, correct, or shut down the system at any point? Are outputs reversible? Is there a meaningful opt-out mechanism?
SCORE LEVELS
What your score means.
80–100
AI Positive Gold
Exceptional ethical performance. The system demonstrably creates significant positive value across all 5 dimensions.
60–79
AI Positive Verified
Strong ethical performance. The system meets high standards on most criteria with documented evidence.
40–59
AI Positive Emerging
The system shows commitment to ethical AI but has measurable gaps. A clear improvement path is provided.
0–39
Not eligible
The system does not yet meet the threshold for AI Positive recognition. Compliance-only status.
IN YOUR FULL REPORT
Where AI Positive appears.
Every Full Report (€690) includes an AI Positive preliminary assessment. This section evaluates your system’s ethical performance indicators based on the information you provide during the intake questionnaire.
The preliminary assessment identifies which of the 5 criteria your system currently addresses, where gaps exist, and what specific actions would improve your score. It is included at no additional cost: because understanding your ethical position is as important as understanding your legal one.
For organizations seeking a dedicated AI Positive audit with in-depth scoring and a verifiable badge, a standalone product is planned for 2027.
COMPLIANCE VS. PERFORMANCE
Two layers. One standard.
| AI Act Compliance | AI Positive™ | |
|---|---|---|
| Objective | Meet legal obligations | Maximize ethical value |
| Framework | EU AI Act (2024/1689) | AI Positive (5 criteria: EU + international standards) |
| Sources | 144-page regulation + implementing acts | EU AI Act + NIST AI RMF + OECD + UNESCO + national frameworks |
| Score type | Risk classification (4 levels) | Performance score (0–100) |
| Question answered | Are you legal? | Are you good? |
| Mandatory | Yes (enforceable) | Voluntary (recommended) |
| Output | Position assessment | Ethical performance rating |
WHAT AI POSITIVE IS NOT
It is not a certification, not yet. It is a framework and a methodology.
It is not a replacement for AI Act compliance. Compliance is a prerequisite.
It is not a marketing label. It is a structured, scored, verifiable assessment.
It does not guarantee ethical outcomes. It measures the conditions that make ethical outcomes more likely.
THE EARLY MOVEMENT
Being first is the advantage.
AI Positive is not just a score. It is a movement. The organizations assessed today are the founding members: documented, dated, verifiable. Their badge carries a date that no one can replicate later.
When hundreds of companies are assessed in 2027 or 2028, the ones who were assessed in 2026 will stand apart. Not because their score is higher, but because their position was established first. In regulatory markets, the date on the document is the proof. Early members of the AI Positive movement own that proof.
This is why the purchase is not about fear. A “minimal risk” result is not a failure. It is a documented clean position with a dated badge. The value is in being assessed, not in having problems. Sprinkling Act has no incentive to inflate your risk. Your position is your position.
INTELLECTUAL PROPERTY
Origin and ownership.
The AI Positive framework, including its 5-criteria scoring methodology, badge levels, and assessment structure, was conceived, designed, and first published by Sprinkling Act (Lamar B. Shucrani, BCE BE 1034.962.482) in March 2026.
The methodology is protected by copyright under EU and Belgian law. The scoring structure, criteria definitions, and assessment process constitute original intellectual work. Any reproduction, adaptation, or commercial use without written authorization is prohibited.
First publication: March 20, 2026. Document integrity anchored on Bitcoin and Ethereum via SHA-256 blockchain timestamp (OriginStamp).
FAQ
About AI Positive
What is AI Positive?
AI Positive is a measurable ethical performance framework for AI systems developed by Sprinkling Act. It evaluates an AI system across 5 criteria (governance, bias reduction, explainability, net human impact, control) for a 100-point score. AI Positive measures positive value creation, not regulatory compliance — compliance under the EU AI Act is the prerequisite.
How is AI Positive different from EU AI Act compliance?
EU AI Act compliance is the legal floor: it tells you what you must do (Articles 5, 6, 50, etc.). AI Positive is the performance ceiling: it tells you how well your AI creates positive value beyond the legal minimum. AI Positive is sequenced after EU AI Act classification — you cannot obtain an AI Positive score without first establishing your regulatory position through a Full Report.
What are the 5 AI Positive criteria?
(C1) Governance & Accountability (0–20), (C2) Bias Reduction & Fairness (0–20), (C3) Explainability & Transparency (0–20), (C4) Net Human Impact (0–20), (C5) Control & Reversibility (0–20). Total: 100 points. Each criterion has documented examples and is assessed against international frameworks (EU AI Act, NIST AI RMF, OECD AI Principles, UNESCO AI Ethics).
What scores qualify for AI Positive recognition?
Four tiers: AI Positive Gold (80–100 points), AI Positive Verified (60–79), AI Positive Emerging (40–59), Not eligible (0–39). Recognition includes a public badge linked to your full assessment, OpenTimestamps-stamped for cryptographic verifiability.
AI Positive assessment is included in every Full Report. Start with the free diagnostic.
SEE ALSO
Article: AI Positive, Beyond Compliance
The full analysis of the AI Positive framework and why compliance is not enough.
AI Agents
The 4 ACTS — regulatory framework for AI agent builders, complementary to AI Positive.
Free Diagnostic
9 questions. Instant classification.
Full Report
15–20 pages with AI Positive assessment included.