Introduction

The report, published in June, 2020, has been prepared by the Montreal AI Ethics Institute, an international non-profit research institute helping people understand the societal impacts of AI and equipping them to take action accordingly. 

In this decade, AI has become the byword for technological progress and is being used extensively across fields, be it combating COVID-19 or spending increasing amount of time online. Taking a note of the development in this field and how it is shaping our society and interactions with each other, the report outlines the most important developments that occurred in the last quarter. The objective is to help readers and organizations to make informed decisions after taking into account the societal impacts of AI-enabled solutions.  

The report covers a range of areas spanning from Agency and Responsibility, Security and Risk, Disinformation, Jobs and Labour to Future of AI Ethics.   

Each of the above head is subdivided into research summaries and articles summaries where both the issues of biasedness as well as the ultimate greater objective and recommendations are discussed. For example, the problem of bias in ranking and recommendation algorithms have been discussed in details. While e-tailers claim to apply the notion of diversity in these algorithms, it ultimately ends up getting stripped of cultural and contextual social meanings. As a solution, the report advocates a framework that does away with rigid, attributed categories and instead builds a probabilistic model based on the data on how a person feels which may contain representations of words, images, and other inputs. 

In another context, the report evaluates the fairness process of machine learning fairness and the lessons learnt from it. To do so, they examined Google’s QuickDraw, an AI system that attempts to guess users’ doodles of items. The goal of the system was to collect data from groups of users by gamifying it and making it freely available online after an image recognition system guesses what it is. However, it was observed that, over time, the system excluded women’s apparel as majority of people drew unisex accessories. The report states that the solution to this will be to think about fairness in a more holistic sense and build communication lines between the user and the product. 

The report also mentions the fact that the rise of AI systems has led to an unintended conflict between economic pursuits like profit making and creating societal benefits from the deployment of these systems. The market incentive structures have mostly harmed morality. However, on the brighter side, in times of discrepancy between ethical and economic incentives, we have the opportunity to steer progress in the right direction. In this context, the author rightly questions the existence and role of humans in the case of rise of superintelligence and hence our actions should be determined by keeping such outcomes in mind. 

Relevance of the Report

The report touches on a wide variety of subject such as the role social media plays in misinformation, the implication of automation on labour and the privacy and effectiveness of COVID-19 apps and the author’s thought process for that subject. It provides the reader with a different angle of viewing the capability of AI which otherwise seemed all-encompassing. In this context it is worth mentioning that when the whole world is of the opinion that AI and ML can bring an end to this pandemic, the report makes a bold statement that AI won’t save us from it. Through this report, the reader is made aware of the various biases that are unknowingly ingrained in AI algorithms and how we can work in a direction to rectify it. 

Key Takeaways

• Progress is the result of conscious human choices which are often shaped and influenced by surrounding social and economic institutions.
• However, in the end, it needs to be understood that the word ethics is too big a word in terms of technology and will be much influenced by the way it is defined. In fact, ethics owners, the broad term for people who are responsible for this within organizations, have a vast panel of responsibilities including communication between the ethics review committees and product design teams. 
• When it comes to AI systems, both users and designers play an active role in shaping those systems’ most significant pieces. 
• Despite the progress achieved, there are many unsolved and open problems that still remains to be addressed for which finding a novel solution is a priority.