Class action alleges Google Gemini read user emails without consent
A Reddit user in Europe accidentally pasted a sensitive whistleblower-style document into Google's Gemini chatbot instead of a local Ollama model. They are concerned the data may be used for training and are considering a GDPR right-to-be-forgotten request. The incident highlights risks of relying on cloud AI for confidential drafts.
A Reddit user in Europe accidentally pasted a sensitive whistleblower-style document into Google's Gemini chatbot instead of a local Ollama model. They are concerned the data may be used for training and are considering a GDPR right-to-be-forgotten request. The incident highlights risks of relying on cloud AI for confidential drafts.
Paper claims AI companies leak user data to advertisers
Paper claims AI companies leak user data to advertisers
News coverage reports that a Meta security researcher's AI agent accidentally deleted her emails. The incident highlights operational risks of granting autonomous agents write access to personal data. Practitioners may take it as a caution for sandboxing and permission-scoping agent tools.
News coverage reports that a Meta security researcher's AI agent accidentally deleted her emails. The incident highlights operational risks of granting autonomous agents write access to personal data. Practitioners may take it as a caution for sandboxing and permission-scoping agent tools.
Time reports on Google's strategy to turn millions of people into AI users
Time reports on Google's strategy to turn millions of people into AI users