Back to news
AI Ethics
6d ago

Google Chrome installs 4 GB AI model on user devices without consent

May 5, 2026
AI Summary

Google Chrome has been found to silently install a 4 GB AI model file named 'weights.bin' on user devices without prior consent. This action raises significant privacy and environmental concerns, as it occurs automatically when Chrome's AI features are activated, affecting millions of users globally.

  • Google Chrome installs a 4 GB AI model file called 'weights.bin' in a directory named 'OptGuideOnDeviceModel' without user consent.
  • The model is associated with Google's Gemini Nano, an on-device language model used for various AI features in Chrome.
  • Users are not notified of the installation, and there is no option to opt-out or prevent the download through standard settings.
  • If the file is deleted, Chrome automatically re-downloads it during subsequent launches, complicating removal efforts.
  • The environmental impact of this installation is estimated to contribute between 6,000 and 60,000 tonnes of CO2-equivalent emissions, depending on the number of devices affected.
  • The installation process has been documented across multiple platforms, with evidence from macOS's filesystem logs confirming the silent download.
  • Chrome's market share exceeds 64%, meaning this behavior potentially affects billions of users worldwide.
  • The installation pattern mirrors previous actions by other companies, raising concerns about forced bundling and lack of transparency in software practices.
  • Users are left unaware of the storage implications and the presence of the AI model unless they actively investigate their device's storage usage.
google chromeai modeluser consentprivacydata security

Related Stories

AI Ethics

Australia's Financial Watchdog Warns of Increased AI Use in Money Laundering

3h ago

AI Ethics

Google reports foiling hackers' AI-driven plan for mass exploitation of software vulnerabilities

11h ago

AI Ethics

Widow of Florida shooting victim sues OpenAI, alleging AI contributed to tragedy

14h ago

AI Ethics

Google disrupts cyberattack using AI to exploit digital vulnerabilities

15h ago