The study’s result seem to be that users often get unique results. That’s not the same as “personalized”, and it certainly isn’t evidence of “bias” as the sprea

The study’s result seem to be that users often get unique results. That’s not the same as “personalized”, and it certainly isn’t evidence of “bias” as the spreadprivacy.org-link suggests. A good faith interpretation would point to google running learning algorithms on their results. That would also seem to be a far better explanation for Google changing parts of the page layout, such as the position of news and video results. The use of the term “bias” for describing differences search results also trips my conspiracy theory detectors.