Pillar 2

Strengthening
the development of Trustworthy AI

MISSION KI Quality Standard

MISSION AI is developing a voluntary quality standard guideline for artificial intelligence (AI) that strengthens the reliability and trustworthiness of AI applications and systems.

Since August 1, 2024, the EU AI Act has regulated AI development and deployment across Europe to promote trustworthy and human-centric AI. The law adopts a risk-based approach, focusing on prohibited and high-risk AI systems. The latter are subject to strict requirements and conformity assessments.

Other AI systems, such as those used in automated contract review or supply chain optimization, are subject to transparency requirements, meaning end users must be informed when interacting with AI. However, the regulation does not directly address the implications of these AI systems in terms of financing or quality management within companies.

This is where MISSION KI comes in. Our initiative aims to create a flexible framework for all AI systems, including those below the high-risk threshold, through a voluntary, practical quality standard. At the same time, the standard aligns with the EU AI Act’s requirements for high-risk AI systems, ensuring compatibility.

MISSION KI enhances market transparency, strengthens AI operators’ trust in AI technologies, and provides a competitive advantage for AI providers applying the standard. By testing real-world applications, we ensure the standard is practical and meets industry needs.

1. What is the foundation of the MISSION KI Quality Standard based on?

The standard is based on the "Ethics Guidelines for Trustworthy AI" developed by the High-Level Expert Group (HLEG) of the European Commission. The HLEG established key principles for assessing AI trustworthiness, which also served as a foundation for the AI Act’s requirements.

1.1 — Values

In this framework, six core values have been identified to guide responsible AI development and deployment.

Reliability

Performance & Robustness
Fallback Plans & General Safety

AI-specific cyber security

Resistance to AI-specific attacks and security

Data quality, protection and management

Data quality & integrity
Protection of personal data
Protection of proprietary data
Data access

Non-discrimination

Avoidance of unjustified distortions
Accessibility and universal design
Stakeholder participation

Transparency

Traceability & documentation
Explainability & interpretability
External communication

Human supervision & control

Human capacity to act
Human supervision

1.2 — Protection needs analysis

The MISSION KI Quality Standard relies on the protection needs analysis (SBA) as a starting point to ensure efficiency. This analysis determines the necessary protection requirements for the defined values and thus forms the basis for a targeted test. It filters out the relevant values and criteria for a use case and defines a target for the subsequent test.

Details of the protection needs analysis

The minimum standard therefore considers the variety of AI application scenarios - from energy distribution optimization and product recommendation systems to medical diagnostic tools. The relevance of the individual values varies depending on the use case.

For example, the value ‘non-discrimination’ plays a subordinate role in an AI for optimizing power distribution, as the decisions are based on technical parameters. In this case, the value of ‘transparency’ takes center stage: the AI's decisions must be comprehensible and understandable so that operators and regulatory authorities can check why certain energy distributions were made.

Regardless of the use case, the ‘reliability’ value is always subject to scrutiny, as it is considered fundamental to the quality of any AI application. The other values can be categorized as not applicable in whole or in part under certain conditions that are clearly defined in the protection requirements analysis.

2. How does the standard become auditable?

2.1 — The test criteria catalogue translates abstract values into measurable variables

In the context of the emerging AI regulation and standardization, a series of (criteria) catalogs and standards on AI trustworthiness have also been published in Europe and Germany. These are largely based on the results of the HLEG-KI.

Details of the test criteria catalogue

The MISSION KI test criteria catalogue is based on three sources in particular:

VDE SPEC 90012,
AI test catalogue of the Fraunhofer IAIS,
AIC4 criteria catalogue for AI cloud services from the Federal Office for Security and Information Technology (BSI).

In order to make the MISSION AI quality standard testable, the 6 abstract values were translated into a structured test procedure based on the so-called ‘VCIO’ approach (Values - Criteria - Indicators - Observables). This is divided into several levels: The values form the foundation on which specific criteria are built. Indicators and measurable variables (observables) are used to assess these criteria. The degree of fulfilment of each value is systematically determined on the basis of this structure. This methodology ensures a precise and comprehensible assessment.

In addition, test tools are developed to check the fulfilment of the observables and thus increase the reliability of the test result.

2.2 — The evaluation

At the end of the test process, an overall assessment is made for each of the six defined values. This assessment is compared with the previously determined protection requirements. An AI application passes the test if it achieves the defined test target for each value. This documents that the quality measures and their evidence sufficiently fulfil the identified protection requirements.

The successful test thus confirms that the AI application fulfills the necessary quality standards and has implemented the required protective measures. This process ensures a thorough evaluation and creates transparency regarding the trustworthiness and security of the tested AI systems.

3. Advantages for AI users and operators

The MISSION AI Quality Standard offers clear advantages for AI providers and AI operators. AI providers benefit from an efficient proof of quality that can be used by large companies and start-ups alike.

This improves their competitiveness, as they can stand out on the market thanks to comparable quality criteria. In addition, with the MISSION AI Quality Standard, AI providers lay the foundation early on to fulfil the requirements of the EU AI Act.

This is particularly helpful if the area of use of your AI application changes in such a way that it is later categorized as a high-risk application. AI operators in turn benefit from greater market transparency and higher reliability of the AI applications used. This also strengthens end users' trust in the technology. A win-win situation that promotes the development of a robust and trustworthy AI ecosystem in Europe.

Frequently Asked Question

Why is the standard voluntary?

The standard is voluntary, meaning it is not legally required but serves as a signal of the quality of an AI system when fulfilled. Compliance with the standard should be motivated by its economic benefits. It allows companies to create market transparency based on uniform guidelines, thus gaining a competitive advantage. Customers and end users also benefit, as a successfully certified AI system becomes more trustworthy. Additionally, the standard serves as a guideline for meeting key regulatory requirements such as the EU AI Act or international standards.

What does "compatible with the AI Regulation" mean?

"Compatible with the AI Regulation" means that the standard aligns with European AI regulations. While it delves deeper into details than abstract regulatory requirements, it still sets guidelines that are substantively linked to legal texts. Furthermore, the MISSION KI Quality Standard`s requirements do not contradict the AI Regulation. Companies can use the standard as a reference to build effective AI compliance management. Since it is compatible with regulations, meeting legal requirements or established standards also facilitates compliance with the MISSION KI Quality Standard. The standard also considers compatibility with sector-specific regulations beyond the AI Regulation.

How do companies benefit from this standard?

Companies benefit in various ways. AI providers enhance their competitiveness and can demonstrate or verify the quality of their AI systems. AI operators, as purchasers of AI systems, gain transparency and can rely on the system’s quality. Reliable and comprehensible development processes are also encouraged. Overall, adherence to the standard strengthens trust in the market.

How is the standard kept up to date?

Artificial intelligence is a highly dynamic and constantly evolving technology, making the standard’s relevance crucial. This is ensured through a modular development approach and sufficiently abstract guidelines. This means that evaluation procedures, description templates, risk assessments, evaluation criteria, testing methods, and tools can be updated separately without compromising the validity of individual components. Additionally, the technically agnostic formulation of the evaluation criteria ensures that new AI methods and developments remain covered by the standard. This makes the standard globally applicable and compatible with evolving AI systems.

What role do ethical principles play in the standard, and how are they practically implemented?

Ethical principles play a central role in the standard. They ensure that AI systems are developed and used responsibly for the benefit of society. The standard is based on the ‘Ethics Guidelines for Trustworthy AI’ of the High-Level Expert Group (HEG-KI)and explicitly includes measures to ensure transparency, explainability, non-discrimination, human oversight, and data protection. Incorporating ethical principles strengthens user and public trust in AI technologies and helps minimize potential risks and negative impacts.

How does the MISSION KI Quality Standard relate to other regulations and standards?

The MISSION KI Quality Standard is closely linked to other regulations and standards. It serves as a practical complement and specification to existing legal requirements, international norms, and evaluation frameworks, such as the EU AI Act, the Fraunhofer AI Evaluation Catalog, VDE SPEC 90012, and ISO/IEC JTC 1/SC 42 standards. The standard helps companies comply with regulatory requirements while ensuring that AI systems are developed and operated in accordance with global best practices and ethical principles. Due to its compatibility with regulations such as the GDPR or MDR, the MISSION KI Quality Standard supports companies in harmonizing their processes and ensuring consistent and trustworthy AI implementation.

What is the difference between high-risk AI and other AI systems in the context of MISSION KI?

High-risk AI is already subject to comprehensive and strict requirements under the AI Regulation. The MISSION KI Quality Standard is not limited to a specific risk category but focuses on practical, concrete requirements that are less extensive than those of the AI Regulation. At the same time, it provides guidance for compliance, as it is aligned with the AI Regulation. Beyond high-risk AI systems, the MISSION KI Quality Standard enables organizations to ensure and transparently communicate the quality of AI systems, even in non-regulated areas. This creates a foundation for trust and transparency.

Who conducts the evaluations, and how is independence ensured?

Evaluations are conducted by either internal or external auditors, depending on the assessment depth. Auditors must have extensive professional experience in developing or assessing AI or IT systems and hold relevant certifications. Independence is ensured by requiring that internal auditors are not involved in the system’s development and must come from an independent department. External auditors must be financially independent from the AI provider and register accordingly.

How does the MISSION KI Quality Standard differ from other international AI standards or guidelines?

MISSION KI stands out for its practical applicability and concrete implementation guidelines for AI systems. It also provides a way for non-regulated AI systems to demonstrate quality and trustworthiness effectively. The evaluation criteria are based on the state of the art and can be applied to various use cases. Additionally, the standard includes a collection of specific testing methods and evaluation tools, distinguishing it from comparable standards.