Apple critique of Google's Topics API based on bad code

Apple has criticized Google's Topics API, which is designed for targeted advertising based on user browsing history, claiming it facilitates digital fingerprinting that could allow advertisers to track users. This critique references a study by researchers from the University of Wisconsin-Madison, which suggests that the Topics API could enable the re-identification of users, even those attempting to maintain anonymity. Apple argues that Google's attempt to mitigate this risk through randomization is insufficient. However, the validity of the research has been challenged by Google engineers, who pointed out a flaw in the study's methodology that significantly inflated the perceived risk of fingerprinting. After correcting the code, the re-identification rate dropped from 57% to around 3%. Despite this reduction, the researchers still assert that fingerprinting is possible, with a potential impact on millions of users. The Topics API was introduced as part of Google's Privacy Sandbox initiative, intended to replace third-party cookies, which have been criticized for privacy violations. Although Google initially planned to eliminate third-party cookies, pushback from advertisers has led to a reconsideration of this approach, allowing both Topics and traditional cookie-based tracking to coexist. Apple emphasizes the importance of developing web technologies that do not exacerbate fingerprinting risks, advocating for a balance between user privacy and web functionality.