← sources

Banks just drew a line on AI agents. Four questions, and I think every ops team gets asked them next.

source: reddit (reddit:AI_Agents)published 2026-10-19 14:13ingested 2026-10-20 02:35

retained text

Same week, two signals.

JPMorgan said outside AI agents can look at customer accounts but can't move money (CNBC, Oct 13). Reuters reported bank regulators are now asking banks four things about their agents:

1. **What is each agent allowed to do?**
2. **Who approved that action?**
3. **Can you trace what it decided?**
4. **Can you switch it off?**

My read: this isn't anti-agent, it's anti-unaccountable-agent, and it won't stay in banking. If you run agents against real systems, your auditor, your security team or your board will ask the same four. Probably this year.

And most fleets can't answer them with evidence. The permissions live in a prompt. The approval was a thumbs-up in Slack. The trace was written by the agent itself. "Switch it off" means killing a process while its credentials keep working.

That's true whether the agent handles refunds, ships code to prod, manages sub-agents, or shops on someone's behalf.

Running agents in production? I'd like to hear where you land:

* **Agree or disagree that these four are coming for every fleet?**
* **Which one would be hardest for you to answer today?**

Not selling anything. Curious whether the people running this see it the same way.

document fetched 2026-10-20 00:19 · canonical https://reddit.com/r/AI_Agents/comments/1x9y040/banks_just_drew_a_line_on_ai_agents_four