智善 · 共济链接人工智能原则平台

8. 公正与公平（Fair and equal）

We aspire to embed the principles of fairness and equality in datasets and algorithms applied in all phases of AI design, implementation, testing and usage – fostering fairness and diversity and avoiding unfair bias both at the input and output levels of AI.

原则: 特利亚电信公司的可信人工智能伦理指导原则（Telia Company Guiding Principles on trusted AI ethics）, Jan 22, 2019

作者：Telia Company AB

相关原则

(d) Justice, equity, and solidarity

AI should contribute to global justice and equal access to the benefits and advantages that AI, robotics and ‘autonomous’ systems can bring. Discriminatory biases in data sets used to train and run AI systems should be prevented or detected, reported and neutralised at the earliest stage possible. We need a concerted global effort towards equal access to ‘autonomous’ technologies and fair distribution of benefits and equal opportunities across and within societies. This includes the formulating of new models of fair distribution and benefit sharing apt to respond to the economic transformations caused by automation, digitalisation and AI, ensuring accessibility to core AI technologies, and facilitating training in STEM and digital disciplines, particularly with respect to disadvantaged regions and societal groups. Vigilance is required with respect to the downside of the detailed and massive data on individuals that accumulates and that will put pressure on the idea of solidarity, e.g. systems of mutual assistance such as in social insurance and healthcare. These processes may undermine social cohesion and give rise to radical individualism.

由 European Group on Ethics in Science and New Technologies, European Commission 在伦理原则和民主先决条件（Ethical principles and democratic prerequisites）发表, Mar 9, 2018

· 4. The Principle of Justice: “Be Fair”

For the purposes of these Guidelines, the principle of justice imparts that the development, use, and regulation of AI systems must be fair. Developers and implementers need to ensure that individuals and minority groups maintain freedom from bias, stigmatisation and discrimination. Additionally, the positives and negatives resulting from AI should be evenly distributed, avoiding to place vulnerable demographics in a position of greater vulnerability and striving for equal opportunity in terms of access to education, goods, services and technology amongst human beings, without discrimination. Justice also means that AI systems must provide users with effective redress if harm occurs, or effective remedy if data practices are no longer aligned with human beings’ individual or collective preferences. Lastly, the principle of justice also commands those developing or implementing AI to be held to high standards of accountability. Humans might benefit from procedures enabling the benchmarking of AI performance with (ethical) expectations.

由 The European Commission’s High-Level Expert Group on Artificial Intelligence 在 “可信赖的人工智能”伦理准则草案（Draft Ethics Guidelines for Trustworthy AI）发表, Dec 18, 2018

2. Good and fair

Data enhanced technologies should be designed and operated in a way throughout their life cycle that respects the rule of law, human rights, civil liberties, and democratic values. These include dignity, autonomy, privacy, data protection, non discrimination, equality, and fairness. Why it matters Algorithmic and machine learning systems evolve through their lifecycle and as such it is important for the systems in place and technologies to be good and fair at the onset, in their data inputs and throughout the life cycle of use. The definitions of good and fair are intentionally broad to allow designers and developers to consider all of the users both directly and indirectly impacted by the deployment of an automated decision making system.

由 Government of Ontario, Canada 在人工智能伦理使用原则[草案]发表, Sept 14, 2023

Ensure fairness

We are fully determined to combat all types of reducible bias in data collection, derivation, and analysis. Our teams are trained to identify and challenge biases in our own decision making and in the data we use to train and test our models. All data sets are evaluated for fairness, possible inclusion of sensitive data and implicitly discriminatory collection models. We execute statistical tests to look for imbalance and skewed datasets and include methods to augment datasets to combat these statistical biases. We pressure test our decisions by performing peer review of model design, execution, and outcomes; this includes peer review of model training and performance metrics. Before a model is graduated from one development stage to the next, a review is conducted with required acceptance criteria. This review includes in sample and out of sample testing to mitigate the risk of model overfitting to the training data, and biased outcomes in production. We subscribe to the principles laid out in the Department of Defense’s AI ethical principles: that AI technologies should be responsible, equitable, traceable, reliable, and governable.

由 Rebelliondefense 在人工智能道德原则发表, January 2023

· 2) Diversity and Fairness:

Artificial intelligence should provide non discriminatory services to various groups of people in accordance with the principles of fairness, equity and inclusion. We aim to initiate from the disciplined approach of system engineering, and to construct the AI system with diverse data and unbiased algorithms, thus improving the fairness of user experience.

由 Youth Work Committee of Shanghai Computer Society 在中国青年科学家2019人工智能创新治理上海宣言（Chinese Young Scientists’ Declaration on the Governance and Innovation of Artificial Intelligence）发表, Aug 29, 2019