· 4. Governance of AI Autonomy (Human oversight)

The correct approach to assuring properties such as safety, accuracy, adaptability, privacy, explicability, compliance with the rule of law and ethical conformity heavily depends on specific details of the AI system, its area of application, its level of impact on individuals, communities or society and its level of autonomy. The level of autonomy results from the use case and the degree of sophistication needed for a task. All other things being equal, the greater degree of autonomy that is given to an AI system, the more extensive testing and stricter governance is required. It must be ensured that AI systems continue to behave as intended when feedback signals become sparser. Depending on the area of application and or the level of impact on individuals, communities or society of the AI system, different levels or instances of governance (incl. human oversight) will be necessary. This is relevant for a large number of AI applications, and more particularly for the use of AI to suggest or take decisions concerning individuals or communities (algorithmic decision support). Good governance of AI autonomy in this respect includes for instance more or earlier human intervention depending on the level of societal impact of the AI system. This also includes the predicament that a user of an AI system, particularly in a work or decision making environment, is allowed to deviate from a path or decision chosen or recommended by the AI system.
Principle: Draft Ethics Guidelines for Trustworthy AI, Dec 18, 2018

Published by The European Commission’s High-Level Expert Group on Artificial Intelligence

Related Principles


When an AI system significantly impacts a person, community, group or environment, there should be a timely process to allow people to challenge the use or output of the AI system. This principle aims to ensure the provision of efficient, accessible mechanisms that allow people to challenge the use or output of an AI system, when that AI system significantly impacts a person, community, group or environment. The definition of the threshold for ‘significant impact’ will depend on the context, impact and application of the AI system in question. Knowing that redress for harm is possible, when things go wrong, is key to ensuring public trust in AI. Particular attention should be paid to vulnerable persons or groups. There should be sufficient access to the information available to the algorithm, and inferences drawn, to make contestability effective. In the case of decisions significantly affecting rights, there should be an effective system of oversight, which makes appropriate use of human judgment.

Published by Department of Industry, Innovation and Science, Australian Government in AI Ethics Principles, Nov 7, 2019


Those responsible for the different phases of the AI system lifecycle should be identifiable and accountable for the outcomes of the AI systems, and human oversight of AI systems should be enabled. This principle aims to acknowledge the relevant organisations' and individuals’ responsibility for the outcomes of the AI systems that they design, develop, deploy and operate. The application of legal principles regarding accountability for AI systems is still developing. Mechanisms should be put in place to ensure responsibility and accountability for AI systems and their outcomes. This includes both before and after their design, development, deployment and operation. The organisation and individual accountable for the decision should be identifiable as necessary. They must consider the appropriate level of human control or oversight for the particular AI system or use case. AI systems that have a significant impact on an individual's rights should be accountable to external review, this includes providing timely, accurate, and complete information for the purposes of independent oversight bodies.

Published by Department of Industry, Innovation and Science, Australian Government in AI Ethics Principles, Nov 7, 2019

· (2) Education

In a society premised on AI, we have to eliminate disparities, divisions, or socially weak people. Therefore, policy makers and managers of the enterprises involved in AI must have an accurate understanding of AI, the knowledge for proper use of AI in society and AI ethics, taking into account the complexity of AI and the possibility that AI can be misused intentionally. The AI user should understand the outline of AI and be educated to utilize it properly because AI is much more complicated than the already developed conventional tools. On the other hand, from the viewpoint of AI’s contributions to society, it is important for the developers of AI to learn about the social sciences, business models, and ethics, including normative awareness of norms and wide range of liberal arts not to mention the basis possibly generated by AI. From the above point of view, it is necessary to establish an educational environment that provides AI literacy according to the following principles, equally to every person. In order to get rid of disparity between people having a good knowledge about AI technology and those being weak in it, opportunities for education such as AI literacy are widely provided in early childhood education and primary and secondary education. The opportunities of learning about AI should be provided for the elderly people as well as workforce generation. Our society needs an education scheme by which anyone should be able to learn AI, mathematics, and data science beyond the boundaries of literature and science. Literacy education provides the following contents: 1) Data used by AI are usually contaminated by bias, 2) AI is easy to generate unwanted bias in its use, and 3) The issues of impartiality, fairness, and privacy protection which are inherent to actual use of AI. In a society in which AI is widely used, the educational environment is expected to change from the current unilateral and uniform teaching style to one that matches the interests and skill level of each individual person. Therefore, the society probably shares the view that the education system will change constantly to the above mentioned education style, regardless of the success experience in the educational system of the past. In education, it is especially important to avoid dropouts. For this, it is desirable to introduce an interactive educational environment which fully utilizes AI technologies and allows students to work together to feel a kind accomplishment. In order to develop such an educational environment, it is desirable that companies and citizens work on their own initiative, not to burden administrations and schools (teachers).

Published by Cabinet Office, Government of Japan in Social Principles of Human-centric AI (Draft), Dec 27, 2018

I. Human agency and oversight

AI systems should support individuals in making better, more informed choices in accordance with their goals. They should act as enablers to a flourishing and equitable society by supporting human agency and fundamental rights, and not decrease, limit or misguide human autonomy. The overall wellbeing of the user should be central to the system's functionality. Human oversight helps ensuring that an AI system does not undermine human autonomy or causes other adverse effects. Depending on the specific AI based system and its application area, the appropriate degrees of control measures, including the adaptability, accuracy and explainability of AI based systems, should be ensured. Oversight may be achieved through governance mechanisms such as ensuring a human in the loop, human on the loop, or human in command approach. It must be ensured that public authorities have the ability to exercise their oversight powers in line with their mandates. All other things being equal, the less oversight a human can exercise over an AI system, the more extensive testing and stricter governance is required.

Published by European Commission in Key requirements for trustworthy AI, Apr 8, 2019

· 8. Robustness

Trustworthy AI requires that algorithms are secure, reliable as well as robust enough to deal with errors or inconsistencies during the design, development, execution, deployment and use phase of the AI system, and to adequately cope with erroneous outcomes. Reliability & Reproducibility. Trustworthiness requires that the accuracy of results can be confirmed and reproduced by independent evaluation. However, the complexity, non determinism and opacity of many AI systems, together with sensitivity to training model building conditions, can make it difficult to reproduce results. Currently there is an increased awareness within the AI research community that reproducibility is a critical requirement in the field. Reproducibility is essential to guarantee that results are consistent across different situations, computational frameworks and input data. The lack of reproducibility can lead to unintended discrimination in AI decisions. Accuracy. Accuracy pertains to an AI’s confidence and ability to correctly classify information into the correct categories, or its ability to make correct predictions, recommendations, or decisions based on data or models. An explicit and well formed development and evaluation process can support, mitigate and correct unintended risks. Resilience to Attack. AI systems, like all software systems, can include vulnerabilities that can allow them to be exploited by adversaries. Hacking is an important case of intentional harm, by which the system will purposefully follow a different course of action than its original purpose. If an AI system is attacked, the data as well as system behaviour can be changed, leading the system to make different decisions, or causing the system to shut down altogether. Systems and or data can also become corrupted, by malicious intention or by exposure to unexpected situations. Poor governance, by which it becomes possible to intentionally or unintentionally tamper with the data, or grant access to the algorithms to unauthorised entities, can also result in discrimination, erroneous decisions, or even physical harm. Fall back plan. A secure AI has safeguards that enable a fall back plan in case of problems with the AI system. In some cases this can mean that the AI system switches from statistical to rule based procedure, in other cases it means that the system asks for a human operator before continuing the action.

Published by The European Commission’s High-Level Expert Group on Artificial Intelligence in Draft Ethics Guidelines for Trustworthy AI, Dec 18, 2018