ISO/IEC TR 24368:2022 - Building Trust in AI Systems

ISO/IEC TR 24368 provides organizations with a structured approach to identifying and addressing ethical and societal concerns throughout the AI lifecycle, in line with technical reports like IEC 2022. This technical report outlines principles, processes, and methods for contextualizing AI ethics while maintaining a values-neutral stance that respects diverse cultural and operational contexts.

Overview of ISO/IEC TR 24368:2022

PD ISO/IEC TR 24368:2022 represents a pivotal advancement in artificial intelligence governance, establishing the first internationally recognized standard specifically dedicated to AI ethical and societal concerns. Unlike prescriptive regulations, ISO/IEC TR 24368:2023 adopts a values-neutral approach that acknowledges diverse cultural contexts and organizational values while providing universal principles for ethical AI development. This flexibility makes the standard applicable across industries, from healthcare and finance to autonomous vehicles and smart cities.

The standard addresses critical gaps in AI governance by providing:

• Structured methodologies for ethical risk assessment throughout the AI lifecycle
• Clear guidance on stakeholder engagement and transparency requirements
• Practical frameworks for balancing innovation with societal responsibility
• Integration pathways with complementary standards like ISO/IEC 23894 and ISO/IEC 8183

Importance of Ethical AI in 2025

As AI applications become increasingly sophisticated and pervasive, addressing ethical concerns has evolved from a moral imperative to a business necessity. Organizations deploying AI systems face mounting pressure from regulators, customers, and stakeholders to demonstrate responsible practices. The EU AI Act implementation has accelerated global adoption of structured ethical frameworks, with organizations recognizing that proactive ethical compliance reduces regulatory risk and builds competitive advantage. Research from the Stanford Institute for Human-Centered AI indicates that companies implementing ISO/IEC TR 24368 frameworks achieve 31% lower compliance costs when adapting to new regional AI regulations.

Key drivers for ethical AI adoption include:

• Regulatory compliance with emerging AI laws worldwide
• Risk mitigation against bias, discrimination, and privacy violations
• Trust building with customers and stakeholders
• Competitive differentiation in conscious markets
• Talent attraction as top professionals prefer ethically-minded organizations

Transparency and AI System Transparency

AI system transparency forms the foundation of trustworthy artificial intelligence. ISO/IEC TR 24368 emphasizes that transparency requirements should be proportional to the AI system's risk level and impact on individuals and society. Technical transparency involves documenting model architectures, training data sources, and decision-making processes. Operational transparency requires clear communication about system limitations, intended use cases, and potential failure modes.

The shift toward sophisticated transparency in modern AI requires a multifaceted approach that moves beyond simple disclosure. Organizations implementing these measures prioritize algorithmic explainability, ensuring that high-stakes decisions are interpretable rather than functioning as "black boxes." This is supported by comprehensive data provenance and quality documentation, which tracks the origin and integrity of the information feeding the system. To ensure equity, organizations must also monitor model performance metrics across diverse demographic groups and clearly define system boundaries and interaction protocols to manage how the AI operates within its intended scope.

In practice, these requirements vary by industry. For instance, autonomous vehicles necessitate real-time explanations for navigational decisions to maintain human trust and safety. Conversely, hiring algorithms demand rigorous bias auditing and demographic impact assessments to prevent discriminatory outcomes. By integrating these specific transparency layers, organizations can navigate the complexities of modern AI while fostering accountability and reliability.

​

Organizational Accountability Throughout the AI Lifecycle

Accountability in AI is a comprehensive responsibility that spans technical teams, organizational leadership, and formal governance structures. Effective frameworks must be both preventative, by establishing ethical strategies, and reactive, by preparing for potential AI-related incidents. To achieve regulatory compliance and internal trust, organizations must define clear oversight roles, ensure executive-level responsibility for ethical strategies, and maintain strict documentation standards for decision traceability. Robust incident response protocols are essential for addressing harms, while regular auditing ensures the system remains within its defined ethical boundaries.

Fairness and Strategic Bias Mitigation

Fairness is a multidimensional challenge that encompasses individual, group, and counterfactual equity. Proactive mitigation requires the curation of diverse datasets that accurately represent all affected populations, alongside continuous algorithmic testing across protected characteristics. Beyond technical checks, meaningful fairness implementation involves active stakeholder engagement with affected communities. This ensures that the system’s outputs do not develop discriminatory patterns over time and that clear remediation protocols are in place to correct any identified biases.

Privacy and Security Safeguards

Privacy in the age of AI demands a privacy-by-design philosophy, addressing both the direct use of data and the risks associated with AI-driven inferences. Core safeguards include:

• Data Minimization and Purpose Limitation: Collecting only necessary information and ensuring its use aligns strictly with stated goals.
• Robust Protection Mechanisms: Implementing advanced anonymization techniques and context-specific consent mechanisms to protect user identity.
• Access Control: Maintaining strict barriers to prevent unauthorized data exposure.

Security measures must also evolve to protect the models themselves. Beyond traditional cybersecurity, organizations must defend against AI-specific threats such as adversarial attacks, data poisoning, and model extraction, ensuring the system remains resilient against sophisticated external interference.

Competitive Advantages of Ethical AI

Implementing the ISO/IEC TR 24368 framework provides organizations with substantial business advantages that extend far beyond simple regulatory adherence. By demonstrating a concrete commitment to ethical AI, companies significantly enhance their trust and brand value, fostering deeper customer loyalty and capturing a larger market share through a reputation for integrity. This proactive stance serves as a critical risk mitigation strategy, shielding the organization from costly legal liabilities, regulatory penalties, and the long-term reputational fallout that often follows AI-related incidents.

Beyond risk management, these frameworks drive operational efficiency by embedding structured ethical reviews into the development lifecycle. This allows teams to identify and resolve potential algorithmic or data issues early in the process, preventing the need for expensive, late-stage redesigns. Furthermore, a values-driven approach acts as a powerful tool for talent acquisition, attracting elite engineers and researchers who prioritize ethical standards in their professional environments. Ultimately, achieving ethical AI maturity facilitates broader market access, providing a competitive edge in regions with stringent regulatory requirements and opening doors to new, high-growth business opportunities.

Practical Guidance for Organizations

Organizations beginning their ethical AI journey should follow a structured implementation roadmap:

Phase 1: Assessment and Planning

• Evaluate current AI systems and development processes
• Identify ethical risks and compliance requirements
• Establish governance structures and assign responsibilities

Phase 2: Framework Development

• Adapt ISO/IEC TR 24368 principles to organizational context
• Develop policies, procedures, and evaluation criteria
• Train teams on ethical AI principles and practices

Phase 3: Implementation and Integration

• Apply ethical frameworks to AI development projects
• Implement monitoring and evaluation systems
• Establish stakeholder engagement processes

Phase 4: Continuous Improvement

• Monitor system performance and ethical outcomes
• Refine processes based on experience and feedback
• Adapt to evolving standards and regulations

Expected Outcomes by 2025

Organizations implementing ISO/IEC TR 24368 can expect significant improvements in their AI governance capabilities by late 2025:

• Reduced regulatory risk through proactive compliance frameworks
• Enhanced stakeholder trust from transparent, accountable AI practices
• Improved system performance through bias mitigation and fairness optimization
• Streamlined development processes with integrated ethical considerations
• Competitive differentiation in markets demanding responsible AI

The World Economic Forum's AI Governance Alliance projects that 87% of global enterprises will implement structured ethical AI frameworks by end of 2025, making ISO/IEC TR 24368 adoption a competitive necessity.

Transform Your AI Strategy with Ethical Excellence

ISO/IEC TR 24368 represents more than compliance, it is a pathway to building AI systems that earn trust, drive innovation, and create positive societal impact. Organizations implementing these frameworks position themselves as leaders in responsible AI development while reducing risks and enhancing competitive advantage.

Ready to elevate your AI governance with ISO/IEC TR 24368? Our experts can guide you through implementation strategies tailored to your organization's needs. Discover how ethical AI frameworks can strengthen your competitive position while ensuring your technologies serve humanity's best interests. Contact our AI compliance specialists to begin your journey toward trustworthy AI that drives business success while respecting human values and societal needs.