NSERC CREATE in Responsible Development of AI (RAI)

来自瑞尔森，麦吉尔，西部，滑铁卢和UQAM的研究人员组成的联盟将开发一个培训计划，使下一代人工智能（AI）专家能够积极了解社会，道德，伦理和政策问题，以及设计有影响力的AI项目的技能，以改善加拿大人的生活。这一定期流CREATE计划的设计充分意识到，人工智能技术可以用于改善福祉，健康，服务和决策的方式，但也可能产生意想不到的副产品，歧视，加剧不平等，侵犯人权，社会排序和破坏民主进程，限制获得服务，并加强对边缘化社区和少数群体的监督和不公平待遇。除了探索人工智能领域的前沿研究外，NSE和非NSE学员还将接受培训，以认识到在纯粹以商业为导向的项目中部署人工智能创新的社会威胁、挑战和负面影响，或者那些由道德不可知论问题解决和/或出于经济利益驱动的项目。培训将涵盖偏见、隐私、透明度、监管、治理、安全和正义以及批判性分析、宣传、知识动员和EDI等问题的理论和实践参与。该计划将创造180多个年度培训机会，涵盖三个跨领域培训流的三个主题研究领域，即人工智能设计道德，隐私增强分析和人工智能问责制。专题研究领域将使学员能够从事尖端的研究工作，并推动人工智能的发展。此外，学员将参与三个培训流，涵盖如何从政府、民间社会和行业等不同角度负责任地开发和部署人工智能。

A consortium of researchers from Ryerson, McGill, Western, Waterloo and UQAM will develop a training program to equip the next generation of Artificial Intelligence (AI) experts with active awareness of societal, moral, ethical, and policy concerns, as well as the skills to design impactful AI projects for social good to improve the lives of Canadians. This regular-stream CREATE program is designed with full awareness that AI technology can be used in ways that improve well-being, health, services and decision-making, but can also have unintended by-products that discriminate, reinforce inequalities, infringe upon human rights, socially sort and disrupt democratic processes, limit access to services and intensify surveillance and unfair treatment of marginalized communities and minority groups. In addition to exploring cutting-edge research in AI, cohorts of NSE and non-NSE trainees will be coached to recognize the societal threats, challenges and negative impacts of deploying AI innovations in purely business-oriented projects or those driven by ethics-agnostic problem solving and/or out of economic self-interest. The training will cover theoretical and hands-on engagement with issues of bias, privacy, transparency, regulation, governance, security, and justice as well as critical analysis, advocacy, knowledge mobilization and EDI. Creating more than 180 annual training opportunities, the program will cover three thematic research areas in three crosscutting training streams, namely AI Ethics by Design, Privacy-enhanced Analytics, and AI Accountability. The thematic research areas will allow the trainees to pursue cutting edge research work and advance the state of the art in AI. In addition, the trainees will be involved in three training streams that cover how AI should be responsibly developed and deployed from alternative perspectives: government, civil society, and industry.