Author: Samantha Booth

In this comprehensive article, we explore the essential steps organisations must take to develop an ethical approach to ensuring great outcomes when using automated systems otherwise known as Artificial Intelligence (AI). Acknowledging the need for ethical AI is the first crucial step, followed by assembling a diverse task force and defining ethical AI principles aligned with the organisation's values. Incorporating ethical considerations at every stage of AI design and development, creating transparent AI systems, and establishing review mechanisms ensure responsible AI usage. Prioritising data privacy and security, training employees on ethical AI use, and engaging with external stakeholders foster accountability and collaborative governance. Additionally, leveraging existing AI frameworks and standards can provide valuable guidance for creating a customized ethical framework. By embracing ethical AI, organisations can navigate the evolving technological landscape responsibly, safeguard against potential risks, and contribute to a more inclusive and prosperous future.

Key steps to take to develop an Ethical AI Approach

Step 1: Acknowledging the Need for Ethical AI

The first step towards becoming ethical AI users is recognising the significance of responsible and ethical considerations in AI implementation. Organisations must acknowledge that AI decisions can significantly impact individuals and communities, making ethical governance a necessity rather than an option. Leaders should educate themselves and their teams about the potential consequences of AI misuse, including biases, privacy violations, and discrimination. Also, the importance of transparency and accountability is all systems and processes.

Step 2: Assemble a Diverse Ethical AI Task Force

To establish a robust ethical AI governance framework, it is essential to create a diverse task force comprising multidisciplinary experts. This team should include data scientists, ethicists, legal professionals, AI specialists, and representatives from impacted communities. Collaboration among diverse stakeholders ensures a comprehensive approach to addressing ethical challenges.

Step 3: Define Ethical AI Principles

The task force must define a set of ethical AI principles that align with the organisation's values and societal norms. These principles should emphasise fairness, transparency, accountability, privacy protection, and avoiding harm to users or society. They serve as a foundation for all AI-related activities and decision-making processes within the organisation.

Step 4: Implement Ethical Design and Development Practices

Ethical considerations should be incorporated into every stage of AI design and development. This includes data collection, algorithm development, and model training. Ensuring diverse and representative datasets and regularly auditing AI systems for biases is crucial. Organisations should encourage responsible innovation while consistently adhering to ethical guidelines.

Step 5: Create Transparent AI Systems

Transparency is vital to building trust in AI applications. Users and stakeholders should have access to clear explanations of how AI systems make decisions. This transparency not only helps in addressing potential biases but also enables users to understand the implications of AI recommendations or actions.

Step 6: Establish Review Mechanisms

Ethical AI governance should include a robust review mechanism to assess AI models and applications for ethical compliance. This could involve regular audits, impact assessments, and user feedback analysis. If issues arise, the organisation must be willing to make necessary changes and improvements promptly.

Step 7: Prioritise Data Privacy and Security

Data privacy and security are fundamental aspects of ethical AI. Organisations must strictly adhere to relevant data protection regulations and ensure that personal data is handled responsibly. Implementing privacy-enhancing technologies and secure data storage practices can help mitigate risks.

Step 8: Train Employees on Ethical AI Use

An organisation's employees play a crucial role in the ethical use of AI. Conducting regular training sessions on ethical AI use, its implications, and best practices helps raise awareness and promotes responsible behaviour among the workforce.

Step 9: Engage with External Stakeholders

Engaging with external stakeholders, including customers, regulators, and advocacy groups, fosters accountability and encourages a collaborative approach to AI governance. Actively seeking feedback and understanding diverse perspectives can lead to continuous improvement.

AI Governance Frameworks

What AI Frameworks exist today

In developing your organisation's own AI ethical framework, you can leverage existing AI ethical frameworks as valuable starting points and guiding references.

Established frameworks have already undergone rigorous evaluations and debates, encompassing a broad spectrum of ethical considerations. By analysing and comparing these frameworks, common themes and core principles that align with the organisation's values and from which goals can be identified. Additionally, these existing frameworks provide insights into potential challenges and concerns that may arise during the implementation of AI technologies. By incorporating their lessons learned and best practices, you can pre-emptively address these issues in your own framework. Building upon these foundations, your organisation can then customise the framework to address specific industry requirements, cultural contexts, and the unique ethical dilemmas that AI applications may pose. By doing so, you not only save time and resources but also benefit from the collective wisdom of the AI community, fostering an inclusive and ethical approach to AI development and deployment within your organisation.

Standards and Certifications

• The IEEE 7000-2021 - IEEE Standard Model Process for Addressing Ethical Concerns during System Design – The goal of this standard is to enable organizations to design systems with explicit consideration of individual and societal ethical values, such as transparency, sustainability, privacy, fairness, and accountability, as well as values typically considered in system engineering, such as efficiency and effectiveness. Projects conforming to IEEE Std 7000 balance management commitments for time and budget constraints with the long-term values of social responsiveness and accountability. To enable this, the commitment of top executives to establish and uphold organizational values is important.
• ‍IEEE CertifAIEd^TM is a certification program for assessing ethics of Autonomous Intelligent Systems (AIS) to help protect, differentiate, and grow product adoption. The resulting certificate and mark demonstrates the organization’s effort to deliver a solution with a more trustworthy AIS experience to their users.

Legislation

• New York City Department of Consumer and Worker Protection to prevent bias (July 23): The city's law requires companies using A.I. software in hiring to notify candidates that an automated system is being used. It also requires companies to have independent auditors check the technology annually for bias. Candidates can request and be told what data is being collected and analyzed. The law applies to companies with workers in New York City, but experts expect it to influence practices nationally.

Principles and Guidelines

• ‍The OECD AI Principles: The OECD AI Principles promote use of AI that is innovative and trustworthy and that respects human rights and democratic values. Adopted in May 2019, they set standards for AI that are meant to be practical and flexible enough to stand the test of time.
• The European Commission's Ethics Guidelines for Trustworthy AI: Published in April 2019, these guidelines propose seven key requirements for Trustworthy AI: human agency and oversight, technical robustness and safety, privacy and data governance, transparency, diversity, non-discrimination, and societal and environmental well-being.
• The United Nations' Principles for the Ethical Use of Artificial Intelligence in the United Nations System: These principles promote the ethical and responsible use of AI to address global challenges and support sustainable development.
• Australia’s AI Ethics Principles: Australia’s 8 Artificial Intelligence (AI) Ethics Principles are designed to ensure AI is safe, secure and reliable.
• The Asilomar AI Principles: Coordinated by FLI and developed at the Beneficial AI 2017 conference, are one of the earliest and most influential sets of AI governance principles.
• U.S. Blueprint For An AI Bill Of Rights (Oct 22): The Blueprint for an AI Bill of Rights is a set of five principles and associated practices to help guide the design, use, and deployment of automated systems to protect the rights of the American public in the age of artificial intelligence.

Proposed Frameworks

Legislation‍

• EU AI Act: The primary A European Strategy for Artificial Intelligence. The proposal sets harmonised rules for the development, placement on the market and use of AI systems in the Union following a proportionate risk-based approach. It proposes a single future-proof definition of AI. Certain particularly harmful AI practices are prohibited as contravening Union values, while specific restrictions and safeguards are proposed in relation to certain uses of remote biometric identification systems for the purpose of law enforcement. The proposal lays down a solid risk methodology to define “high-risk” AI systems.

Industry Initiatives

• Global Initiative on Ethics of Autonomous and Intelligent Systems: The IEEE has several initiatives and guidelines focused on ethical AI, including the "Ethically Aligned Design" document series.
• The IBM AI Ethics Framework: Developed by IBM, this framework centres on four principles: fairness, explainability, robustness, and accountability.
• The Microsoft Responsible AI Standard : Microsoft's framework focuses on designing AI systems that are fair, reliable, private, secure, and inclusive.
• The Partnership on AI's Tenets: This organization brings together various stakeholders to collaborate on AI ethics and development. Their tenets emphasise fairness, transparency, privacy, and accountability.

These are just some examples of AI ethical frameworks that have been proposed by different organizations and institutions. The common goal of these frameworks is to guide the responsible development and deployment of AI technologies, considering the potential impact on individuals, societies, and the environment.

Transitioning from using AI without governance to becoming ethical AI users is a progressive journey that requires dedication, collaboration, and ongoing efforts. By acknowledging the need for ethical considerations, building transparent and accountable systems, and involving diverse stakeholders, organisations can ensure the responsible and beneficial use of AI in a rapidly advancing technological landscape. Embracing ethical AI not only safeguards against potential risks but also paves the way for a more inclusive and prosperous future.

‍

‍

‍