Wise's answer to agent payments: agree once, then the agent acts within terms
AI Engineer · youtube · 2026-11-05
In an AI Engineer talk, Emil Chetty (who leads data and AI product at Wise) argues AI assistants have turned customers into an approval queue. Wise moved $243B across borders last year for 19M customers under 80+ licenses, so every payment needs proof of authorization. His pattern: the customer agrees to the whole job once as a locked agreement; the agent picks the steps; Wise's existing payment software enforces the terms — so even an agent tricked via prompt injection can't pay a new recipient. Covers regulator expectations, retries and changed quotes, the implementation, and how Wise can answer for agent actions.
More from coding & agent
- Running Claude Code from a phone: 60% of a day's work done in 30 minutes — PrajwalTomar_ · 2026-11-05
- OpenClaw robotic arm works as release manager, cherry-picking thousands of PRs — vincent_koc · 2026-11-05
- One prompt, one playable 3D cyberpunk racing game: hands-on with Tencent's Hy4 — HeyAmit_ · 2026-11-05
- Tim O'Reilly at AgentCon: open architecture beats open-source licenses for the agentic web — bibryam · 2026-11-05
- Dev tip: install GitHub CLI for agent dev so agents can see issues, PRs and CI — DanielLockyer · 2026-11-05
- Compacting with the same model cuts Claude Code session costs by ~75% via cache hits — JoshuaJosephson · 2026-11-05