AI connector value lies in secrets store and personal context, not payments
jeff_weinstein · x · 2026-09-16
In a thread on ChatGPT's link connector, engineers argue its real value is less about payments and more about being a trusted, multi-agent/multi-platform secrets and access layer.
Key points:
- Users hold many secrets (starting with passwords) they'd want AI apps to use safely; connectors are the natural vehicle (a 1Password connector is one candidate)
- What Google should build: a personal brain. Granting raw Gmail API access forces too much trust and is inefficient and unreliable versus structured personal context
- Imagine AI shopping with access to a wiki of your purchases, reasons for choices, and preferences — personalization would improve dramatically
More from coding & agent
- Sakana Marlin adds Interactive Reading and PowerPoint export for deep research reports — SakanaAILabs · 2026-09-16
- Andy Matuschak demos programmable highlighters that trigger AI actions on annotations — andy_matuschak · 2026-09-16
- Dev shares trick: dedicated subagent filters SSH output to protect agent context — SaleemUsama · 2026-09-16
- Raw Markdown Folders Kill Your Agent's Context Window: Lessons From 100k Installs — Alone-Biscotti6145 · 2026-09-16
- Nous Research Used 1,393 Subagents to Cut a 1M-Line Codebase by 34.4% in 19 Hours — NousResearch · 2026-09-16
- One AI system for 8 months: don't chase new tools, absorb them incrementally — evielync · 2026-09-16