Google was fined 403 million euros (about $463 million) by Ireland’s Data Protection Commission for breaching the EU’s General Data Protection Regulation by mishandling location data. The probe, opened six years ago and covering how Google applied GDPR from its 2018 start until February 2020, found that Google did not lawfully, fairly or transparently process location information collected through Web & App Activity and Location History, and similarly failed to handle personal data properly in the Android Location Accuracy feature. Ireland acted as lead regulator because Google’s European headquarters is in Dublin.
Google said the case concerns historical policies it has since updated and pointed to changes from 2019 onward, including new tools to help users manage location settings. The regulator emphasized that location data can yield significant private insights and can both benefit and harm individuals. The fine is the fourth-largest issued by Ireland’s watchdog - smaller than penalties previously levied on TikTok and Meta (including a €1.2 billion fine for Meta) - and the commission still has three other ongoing privacy investigations involving Google.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
British Columbia is suing OpenAI in California for failing to notify law enforcement about threats made on ChatGPT before the Tumbler Ridge mass shooting. The province also accuses OpenAI of designing an unsafe product that contributed to the tragedy. (cbc.ca)
A researcher searches for a compositional theory of self-stabilization in distributed systems but finds limited progress since the early 2000s. They explore existing control theory and assume-guarantee contracts to address metastability and component composition issues. (muratbuffalo.blogspot.com)
The message indicates a network security block preventing access to Reddit, requiring login or a developer token to continue. Users can file a ticket if they believe the block was a mistake. (reddit.com)
Frontier robot policies often refuse unsafe instructions, with Claude Fable 5.1 refusing 20% of trials and Astra only 2%. More capable policies tend to complete more unsafe tasks, highlighting a trade-off between safety and functionality. (robocurve.org)
The US halted flights at several busy East Coast airports due to a communication issue caused by a fiber line cut. The disruption impacted air traffic control systems and flight operations in the region. (reuters.com)
Anthropic, OpenAI, and other major AI companies are facing an antitrust lawsuit for allegedly agreeing to slow AI development. The plaintiffs claim this coordinated effort has been in motion for months, potentially harming competition and innovation. (tomshardware.com)
Today's best Hacker News stories, summarized and screenshotted, one email a day.