Search

News reports in 2016 suggested that a leading search engine was suppressing negative search suggestions for one US Presidential candidate but not for her opponent. We conducted a progressive series of five experiments to determine what effect differential suppression of this type might have on voters. We found that negative suggestions attract far more clicks than neutral or positive ones, consistent with extensive research on negativity bias, and that the differential suppression of negative search suggestions can turn a 50/50 split among undecided voters into more than a 90/10 split favoring the candidate for whom negative search suggestions were suppressed. We conclude that differentially suppressing negative search suggestions can have a dramatic impact on the opinions and voting preferences of undecided voters, potentially shifting a large number of votes without people knowing and without leaving a paper trail for authorities to trace. [ABSTRACT FROM AUTHOR]

This article tracks developments at the national level in key European countries in the area of IT and communications and provides a concise alerting service of important national developments. It is co-ordinated by Herbert Smith Freehills LLP and contributed to by firms across Europe. This column provides a concise alerting service of important national developments in key European countries. Part of its purpose is to complement the Journal's feature articles and briefing notes by keeping readers abreast of what is currently happening "on the ground" at a national level in implementing EU level legislation and international conventions and treaties. Where an item of European National News is of particular significance, CLSR may also cover it in more detail in the current or a subsequent edition. [ABSTRACT FROM AUTHOR]

• Search engine autocomplete algorithm bias is consistent with societal discrimination. • Disadvantaged groups face stronger algorithmic bias in autocomplete predictions. • The degree of algorithmic bias varies across topic categories of search terms. • Search engine mediatization, mechanisms & consequences of autocomplete bias discussed. This paper examines the autocomplete algorithmic bias of leading search engines against three sensitive attributes: gender, race, and sexual orientation. By simulating search query prefixes and calling search engine APIs, 106,896 autocomplete predictions were collected, and their semantic toxicity scores as measures of negative algorithmic bias were computed based on machine learning models. Results indicate that search engine autocomplete algorithmic bias is overall consistent with long-standing societal discrimination. Historically disadvantaged groups such as the female, the Black, and the homosexual suffer higher levels of negative algorithmic bias. Moreover, the degree of algorithmic bias varies across topic categories. Implications about the search engine mediatization, mechanisms and consequences of autocomplete algorithmic bias are discussed. [ABSTRACT FROM AUTHOR]