Testing proactive AI agents on my email and calendar: dot caught a date mixup, Muse is noisy
hazelcough · x · 2026-10-10
The author gave several AI agents read access to their personal email and calendar to observe unprompted proactive behavior, ranking: useful proactivity > silence >>> noisy proactivity.
- dot ✅: Cross-referenced a doctor's appointment confirmation email against the calendar and caught that the appointment was on the wrong day, correctly deciding what merited a push notification.
- Muse ❌: Sent a redundant notification about a stranger tagging the author's @.michelle handle on someone else's GitHub gist; it surfaced the same email but missed the date mismatch.
- GrokBot 🟡: Hasn't notified anything — silence beats Muse's noise.
Takeaway: good proactivity means cross-checking multiple sources before interrupting the user.
More from coding & agent
- Voyager's Next Release Will Let You Build Playable Games on X — ravisparikh · 2026-10-10
- "Claude Code is the greatest video game of all time," quips developer — jarrodwatts · 2026-10-10
- Alchemy + Cloudflare deploy praised as unbeatable feedback loop for agent dev — samgoodwin89 · 2026-10-10
- Over half of Vercel's deployments now start with coding agents; Supabase adds 1M databases a week — FinanceYF5 · 2026-10-10
- Learning math with ChatGPT as tutor: paper solving plus a scanning camera — generativist · 2026-10-10
- Dev Builds Robot Face Assembly Entirely with Codex, Awaits ESP32-S3 Shipment — kamathsblog · 2026-10-10