The AI notetaker risk is now an everywhere problem

Summary

The article argues that AI notetakers now record from wearables and personal devices, not only from meeting platforms, so no bot announces itself. It cites Granola's Apple Watch app launched July 28, DEF CON banning smart glasses, and Meta's acquisition of Limitless, and notes that recording legality varies by US state. It also reports that notetakers can be manipulated through prompt injection and early-join narration and can carry context from prior meetings into shared summaries.

Related items

• InfoSocial-HUMINT: An LLM-Augmented Unified Framework for Measuring Social Media ExposureSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoAgentPhishing: A Benchmark of LLM Email Agents Under Adaptive AttackSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoAfterDeleteBench: Auditing Post-Deletion Information Persistence in LLM AgentsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• HighCVE-2026-111079: Goose sends unsanitized terminal history to LLM provider endpointsSimilar attack · NVD/CVE Database
• InfoIs Claude Code safe for the enterprise? What the official settings cover, and what they don'tSimilar attack · Adversa AI Blog