Welcome to the EI weekly round-up; a curation of quality posts to help you cut through the noise and get right to the heart of the discussion on AI and Tech Ethics.
Every Tuesday we publish a list of links to articles and debates that have happened over the past week in the community, allowing you to stay as up-to-date as possible on developments and facts. We will often link to arguments from all sides of the debate, even if the opinions may be controversial. We would like to mention, however, that EI does not endorse any of the information published, all links are reflections of the author's opinions and not that of Ethical Intelligence.
The Second Wave of Algorithmic Accountability
"... the first wave of algorithmic accountability focuses on improving existing systems, a second wave of research has asked whether they should be used at all—and, if so, who gets to govern them".
Frank Pasquale argues that we can distinguish the "... first wave of algorithmic accountability research and activism", which has targeted existing systems and helped illuminate urgent ethical concerns in the AI systems already online, from "...an emerging “second wave” of algorithmic accountability has begun to address more structural concerns.".
"Both waves will be essential to ensure a fairer, and more genuinely emancipatory, political economy of technology".
Machine Learning on Encrypted Data Without Decrypting It
"Recent breakthroughs in cryptography have made it practical to perform computation on data without ever decrypting it. In our example, the user would send encrypted data (e.g. images) to the cloud API, which would run the machine learning model and then return the encrypted answer. Nowhere was the user data decrypted and in particular the cloud provider does not have access to either the orignal image nor is it able to decrypt the prediction it computed."
This application of homomorphic encryption systems might mitigate a number of data protection problems, even though it would still be imperative to ensure the data was collected ethically, and that the data itself is free from biases that are not understood and accounted for.
Tainted Data Can Teach Algorithms the Wrong Lessons
Security problems become ethics problems when vulnerabilities in software systems produce risks in their application that the stakeholders (both users, and the people in the social environment) are unable to understand, assess, and freely and knowingly accept.
Adversarial input is a particularly powerful way to undermine machine learning systems and to cause them to behave in unexpected and unintended ways. "“Current deep-learning systems are very vulnerable to a variety of attacks, and the rush to deploy the technology in the real world is deeply concerning,” says Cristiano Giuffrida, an assistant professor at VU Amsterdam who studies computer security, and who previously discovered a major flaw with Intel chips affecting millions of computers."
In a recent paper Luciano Floridi draws our attention to the extension of the practice of ethics dumping, "the export of unethical research practices to countries where there are weaker legal and ethical frameworks" into the digital realm. This is an ethical risk, but it is also a more basic risk to the quality of research. This article argues that there are also security problems with this sort of practice. "... some companies outsource the training of their AI systems, a practice known as machine learning as a service. This makes it far harder to guarantee that an algorithm has been developed securely."
Algorithms, Automation, and News
A special issue of the Journal of Digital Journalism has been published on "Algorithms, Automation, and News".
Against Ethical AI: Guidelines and Self Interest
"In this paper we use the EU guidelines on ethical AI, and the responses to it, as a starting point to discuss the problems with our community's focus on such manifestos, principles, and sets of guidelines. We cover how industry and academia are at times complicit in ‘Ethics Washing’, how developing guidelines carries the risk of diluting our rights in practice, and downplaying the role of our own self interest. We conclude by discussing briefly the role of technical practice in ethics."
5 Q’s for Anne Kao, Senior Technical Fellow at Boeing Research and Technology](https://www.datainnovation.org/2019/11/5-qs-for-anne-kao-senior-technical-fellow-at-boeing-research-and-technology/)
"The Center for Data Innovation spoke with Anne Kao, Senior Technical Fellow at Boeing Research and Technology. Kao discussed how she uses machine learning to analyze maintenance reports and how philosophy influences how she approaches data science."
Ukraine denounces Apple for calling Crimea part of Russia in apps](https://www.reuters.com/article/us-apple-ukraine-crimea-idUSKBN1Y124O)
"Reuters reporters in Moscow who typed the name of the Crimean provincial capital Simferopol into Apple’s Maps and Weather apps on Wednesday saw it displayed as “Simferopol, Crimea, Russia”. Users elsewhere — including in Ukraine’s capital Kiev and in Crimea itself — see locations in Crimea displayed without specifying which country they belong to. "
One might wonder if the technical specifications in the ticket for the engineering work that was involved discussed the political and ethical implications. It's perhaps difficult to imagine that something of this magnitude wasn't remarked upon, but on the other, so much engineering work in software proceeds as though it happens in at least partial isolation from the downstream social and political environment.